| WIKI | FAQ | Tech FAQ | |
 http://csua.com/feed/ | |
| ||||||
| 11/23 |
| 2008/11/14-26 [Computer/SW/OS/Linux, Computer/SW/OS/Solaris] UID:51989 Activity:moderate |
11/14 lulz why doesn't GOOG buy JAVA i mean SUN i mean whatever the hell they
are these days.
\_ Even GOOG isn't THAT stupid
\_ Sorry, but WHY would Google do something like that? They
run 99.2% Linux servers on the backend. They don't use
Solaris for development. I mean, what does Sun have to
offer to anyone these days?
\_ ZFS, some SMP goodness, a quality OS, some neat stuff
like containers, Java, and MySQL but I'll admit it's not
much which is why the price is where it is. Sun has more
to offer than Apple does in terms of technology, but
can't seem to connect with users the way Apple does.
\_ Apple is a consumer electronics company, not a
technology company. Sun's "quality" OS is being
phased out virtually everywhere it's implemented. -tom
\_ It is true that Solaris is dying, but it is not because
it is inferior.
\_ So? It's proprietary, it's slow-moving, it's
expensive. If they'd community-sourced it 10 years
ago it might have beat out Linux, but at this point
it's dead. -tom
\_ It's not proprietary nor expensive. I'm not
sure what slow-moving means in this context.
\_ It's not proprietary? I can release
"Tom's Kewl Solaris Distribution," and
mirror all the patches Sun puts out? Don't
think so. Slow-moving means it's slow to
support new hardware, it's slow to get vendor
support for commercial/propietary applications,
it has poor support from most open source
packages as well. -tom
\_ People do that with Redhat all the time.
\_ Google is fundamentally an advertisement company. Sun is a
computer hardware company. What kind of "synergy" you are thinking
again?
\_ I would say Sun is a software company at this point.
\_ I work at the software side of Sun. and don't I wish
Sun is a software company. But consider that almost 90% of
the revenue is coming from hardware sales, I would consider
it is still a hardware company more than anything else.
\_ And yet it's not, because SPARC is all-but-dead. The
only value-added Sun has is in software. They better
figure that out right quick.
\_ This is similar to the problem Apple faced in
the mid 90s. People were saying they needed to
license the OS to grow the market share for the
platform, but 80%+ of company revenues were from
hardware. They licensed the OS, got taken to
the cleaners by clone makers, and tanked badly
while failing to grow market share. Jobs came
in and focused the company on hardware, found an
effective niche and has been quite successful.
If Sun were to transition from being a hardware
company to being a software company, they'd have
to be prepared to cut 50%+ of their workforce,
and make a strong case for why the new software
company is something that businesses should be
investing in, which will be difficult since
businesses are currently deciding to get rid of
Sun software. Sun actually has some decent
commodity Intel servers these days, but that's
not going to save them. -tom |
| 2008/11/14-26 [Computer/SW/Languages/Java, Computer/SW/OS/Solaris] UID:51970 Activity:moderate |
11/13 http://sfgate.com/cgi-bin/article.cgi?f=/n/a/2008/11/14/financial/f051352S72.DTL http://preview.tinyurl.com/6nngpm Sun Microsystems Inc. plans to cut up to 6,000 jobs, or 18 percent of its global work force, as sales of its high-end computer servers have collapsed. The drastic move announced Friday highlights Sun's desperation to cut costs and survive as an independent company. Sun's shares have fallen so steeply they've crossed an ominous threshold, driving the company's market value below its cash on hand. That means investors believe the company itself is essentially worthless. lulz is this because of open source and linuz? \_ If Sun goes down, what happens to Mysql? \_ Who cares? \_ I care a little. You may mock Mysql, but it is used by freakin' everyone these days. \_ Someone will buy Sun. heck if they're worthless, maybe I can buy Sun. I've been refraining from double lattes. \_ Sun has become DEC. It was a sad day when Compaq bought DEC. It will be a sad day when Dell or its equivalent buys Sun. -ex-Sun \_ Is this because they opensources Solaris? How do opensource companies make money again? \_ it has nothing to do with Solaris being open. \_ I am a Sun guy. I guess I am on the software-side of the house, so things are not as bad as the headline says. \_ IBM could buy Sun right now outright, lay everyone off, keep the contracts, and pay for the acquisition with the cash inside of Sun. "lulz" \_ shut up paolo \_ Bush is responsible for state of current economy! Free Tibet \_ shut up emarkp \_ You forgot to add "Iraq War" and "lolz", troll. \_ I am a Sun guy. I guess I am on the software-side of the house, so things are not as bad as the headline says. \_ As a ex-Sun employee I am not surprised. Sun has been mismanaged for 10+ years, and is full of deadwood and useless middle managers. Even this job cut won't be enough for the company to survive. Pony tail boy needs to reduce the work force down to about 10K and put an end to the java religion w/in the company is to survive in the long run. \_ As a ex-Sun employee, I think there are many reasons for Sun's problems. Linux is one reason, at least in the workstation / low-end server market. Intel is another reason. Sparc is just not all that important anymore. But the two biggest reasons I think are: (1) extremely poor management; and (2) java. Management at the upper levels was always unwilling to see reality and did not make the cuts that were needed in the early part of this decade. If Sun had cut its staff to 10K-15K in 2002-2003, they would be reasonably well positioned today. Also, Sun has way too many middle managers and upper level technical deadwood (architects, sr. staff eng., &c.). This was a problem that management was also unwilling to correct. These people are the ones that foster the java-religion w/in the company. It doesn't matter how lousy a java project is, it will always be selected for funding over a non-java project. In fact, I saw profitable non-java projects cancelled in favor of incomplete and unreleasable java projects. The only reason for these cancellations was that the profitable projects weren't using java. \_ Religion? You mean cult. -I hate Java \_ http://finance.yahoo.com/q/bs?s=JAVA Cash on hand is 2B, market cap is 3B, so your first premise is incorrect. Also, what you should really be looking at is real tangable assets minus liabilities and that is even less, more like $1B. But JAVA really is cheap. \_ I disagree. Take a look at: http://finance.yahoo.com/q/is?s=JAVA http://finance.yahoo.com/q/is?s=JAVA&annual Maybe I am not reading this right, but it looks like Sun has lost about $1.3 billion to date this year, and lost $864 million in 2006. This trend is not new and reflects the unwillingness of Sun management to face the reality that fewer and fewer customers need sparc, java, zfs, &c. and that too many people are employed in developing things that no one wants to buy. - ex-Sun that no one wants to buy. -ex-Sun \_ Look at the cash flow chart: http://finance.yahoo.com/q/cf?s=JAVA&annual They had positive cash flow in both 2006 and 2007, and while 2008 cash flow is negative (through June), cash flow from operations was positive; the negative hit is on sale purchase of stock, which probably means they spent some cash to do a share buyback. (Which was probably a mistake, looking at the overall situation). In general, they're not bleeding money; they're just becoming less and less relevant. -tom \_ They are burning through $250M per quarter, which means they have a year left if things don't turn around quickly. \_ Actually, it looks they just need to stop buying their own stock. \_ Ponytail can do no wrong. Really does anyone take seriously a man in a ponytail? \_ My Little Pony |
| 2008/9/24-29 [Computer/SW/OS/Windows, Computer/SW/OS/Solaris] UID:51283 Activity:nil |
9/24 Why is nscd going crazy? DoS?
\- back in the solaris say 2.5-2.6 era, it had both some bugs
(some malformed nis maps made it go crazy) and architectural
flaws in the IPC/door+threading mechanism. if you are running
OS-recent, dunno, but you can trace it.
\_ Yeah, I think it's just buggy. I've restarted it, and it seems
to be working again for now. --mconst |
| 2008/4/3-9 [Computer/SW/Languages/Misc, Computer/SW/OS/Solaris] UID:49658 Activity:nil |
4/3 Solaris experts: I've never played with ZFS. Does it have a native
dump command a la ufsdump?
\_ This might be what you are looking for:
http://preview.tinyurl.com/2xqkda [sun - bigadmin] |
| 2008/3/30-4/6 [Computer/SW/OS/Solaris] UID:49614 Activity:nil |
3/30 Question: I just deleted 60 GB of files from an 80 GB disk. The
disk activity lights were blinking like crazy and I could hear the
drive crunch while the data was deleted. This is under Solaris.
Anyway, I think UNIX uses unlink() when files are deleted. Shouldn't
it just update the free list on the superblock and call it a day?
What is all the crunching about?
\_ Well, I guess it depends on how you delete it. I assume you did
'rm -rf *'? In that case it would go through all the directories
delete each file by name one at a time. You could get the behavior
you were hoping for by formatting or rewriting the partition table. |
| 11/23 |
| 2007/11/27-30 [Computer/SW/Languages/C_Cplusplus, Computer/SW/OS/Solaris] UID:48701 Activity:high |
11/27 I'm using select to do a nonblocking check to see if a single socket
has anything to read off it. Problem is, I can have up to 12228
file descriptors, and Linux fd_set only supports up to 4096. Any idea
what I can do about this? (Or a better solution?) -jrleek
\- 1. who are you
2. i am busy this week and you didnt mention language
[i am not fmailar with stuff like java nio] but you might
look at this ucb/cs paper ... matt welsh et al "a design
framework for highly scalable systems" as well as
some of the discussion around libevent.
see the links and graph at http://monkey.org/~provos/libevent
\_ Ah, the program is all in 'C', but it needs to run on multiple
Unix variants. -jrleek
\_ Have you profiled it? Can you port to python or another
scripting language with reasonable performance (alas, not
ruby at this time)? -dans
ruby at this time)? At http://Slide.com, a hot startup in
downtown San Francisco (we're hiring!), we open AND
close millions of socket connections every day. -dans
\_ How could I profile it? This isn't a webserver, it's a
server that accepts, and acts on, messages based on a
protocol I wrote. (Over TCP). In this case, I need to
know about the performance of a tiny part of the code.
I'm not sure how to get that information. gprof, for
example, doesn't seem to allow me to choose just a small
section to profile, and lacks the necessary resoluton
anyway. -jrleek
\_ What you want is a profiling tool that doesn't work
via random sampling but that lets you add profiling
hooks into your code. I've written some homegrown
things like this in the past to profile very tight
loops in massive projects, but I'm sure there are
plenty of better tools out there if you poke around.
\_ Hint: leek >> dans. Why are you listening to his
babbling?
\_ Well, I have a lot to learn about network
programming, so I'll take what I can get. Thanks
for the compliment though. -jrleek
\_ Alas, I don't have any better suggestion that
gprof, though there must be better tools out
there. Another alternative would be to compile
the source, look at the ASM output and try to
hand optimize. Consider that a WAY last
resort, and not worth pursuing unless you're
already a fan of ASM. Two problems though:
there. Another alternative would be to
compile the source, look at the ASM output
and try to hand optimize. Consider that a
WAY last resort, and not worth pursuing unless
you're already a fan of ASM. Two problems
though:
a) it doesn't really scale if you need to
target multiple platforms and b) it's actually
tough to beat a *good* modern optimizing
compiler even if you really know what you're
doing. -dans
\_ UPDATE: I just asked one of the guru's and
he responded, 'books? what are those' see:
http://www.kegel.com/c10k.html
It's more of a jumping off point, but it will
at least give you tools to work with and
references potential implementations. -dans
\_ Hint: In the last three years have you...
a) worked on a project with me?
b) read or hacked any code I've written?
c) used a service based on my code or systems I
administered?
Unless you can answer yes to at least two, you
have NO FUCKING IDEA WHAT YOU'RE TALKING ABOUT.
Why two? Because I built the systems half (as
opposed to the network/routing half) of the
anycast DNS rig that runs the roots for over
fifty ccTLD's including, amusingly enough, .cx.
Thus, the answer to c) is almost always yes.
-dans
\_ Use poll() instead of select, or do multiple selects with several
different fd_sets . -ERic
\_ Can you increase the max size of fd_set in /proc? I'm guessing not,
but couldn't hurt to look. Also, using select on that many file
descriptors will probably result in sucky performance. -dans
\_ Do you know where I can read up on getting really good
performance out of the POSIX tcp codes?
\_ I wish I did. Most of what I know is a collection of voodoo
and lore. It's not super complicated, basically you want to
use non-blocking sockets and poll. Also, avoid threads
unless you know what you're doing. Writing correct threaded
code is hard, writing high-performance threaded code is even
harder. On Linux, processes are basically threads, but with
processes you don't have to handle any locking crap. -dans
\_ epoll (linux) or kqueue (bsd)
\_ Unfortunately, it needs to run on AIX (IBM's Unix) as well.
\- arent you that fellow at livermore? if this is going to
run on ibm big iron, maybe if you have a "user services"
group they will know this. cray and ibm have some people
group they will know this. cray and ibm have had some people
stationed here as part of nersc. i am familar with assos,
fleebsd, and solaris [/dev/poll] but not aix. btw some of
the select vs poll people seem to be unaware of many
places where the interface is different, but under the hood
they are the same thing. --psb
fleebsd, and solaris [/dev/poll] but not aix. --psb
\- i have never heard of/used this [i no longer
work on aix] but check this out:
http://tinyurl.com/26z9jf [pollset]
often i would think if there was something
that was obscure it probably wouldnt be that
good, but in this case 1. ibm has a history
of sitting on good things that fail due to
obscurity 2. i'm not in the loop [no pun
intended] any more on ibm stuff --fmr ibm person
\_ We do have such a group, but they don't know much about
TCP. It's kind of an odd thing to be doing. I wrote this
TCP implementation as a proof-of-concept, but we've never
gotten to money to do something better, so I just keep
trying to improve it incrementally. -jrleek |
| 2007/7/9-12 [Computer/SW/OS/Solaris, Computer/HW/CPU] UID:47240 Activity:high |
7/9 Are 64-bit Linux servers popular compared to 32-bit ones? My company
has a server product that supports 32-bit, and we're trying to see if
it's worth supporting 64-bit as well. Thanks.
\_ in my company, everything linux is 64bit except the handful of
redhat satellite servers, because redhat's sat server product
doesn't support running on 64bit systems (yet). DOH!
\_ Compared to 32bit? No. Whether it is worth it to support or not
depends on a lot of things, but mostly the desires of your
customers and the cost of doing supporting it.
customers and the cost of supporting it.
\_ It has more bits so it must be better! Seriously though, now that
64 bit is here I think you'll find most places doing 64 bit. The
best people to ask are your current customers. Unless you're
making a pre-packaged toaster product, then it doesn't matter.
\_ No, you will not find most places doing 64bit. It's still early.
\_ Maybe at your place. See the person below for an exmaple.
\_ I went to USENIX recently and 64bit is still far from
the norm.
\_ No 64bit at all in most of the places of USENIX
attendees? You know this how?
\_ Did I say: "No 64bit at all?" No, I did not. However,
the topic came up in a session and only a few
people said they used it in production.
\_ Representing how many servers and which companies?
Was this the "64 bit linux" session?
\_ How many are represented by you and two other
MOTD trolls?
\_ Oh, ah, personal attack! Way to go! You
are right, 64 bit is DOA. How many? Like
the other poster we have been building only
64 bit boxes in the last few months and
have the remaining non-64 bit scheduled for
termination over the coming months. Maybe
your place of business is just too small or
doesn't do real work?
\_ I never said it was DOA. It will
gradually become the standard.
However, it is currently not the
standard. Even you admit that you
only started to switch over a few months
ago.
\_ You trolled. I sarcastically trolled
back with the DOA overstatement. I
admit nothing. I told you we've been
at it for a few months which started
last year and continues forward as we
have time for it and new machines come
online. I expect most places will be
doing similar gradual rollouts. The
idea that a few usenix attendees at
a single session is representative of
industry is no more likely than your
'motd trolls' as you call us are
representative of industry. Less so
since we're actually here to discuss
it with you, not a vague 'please
raise your hand' at a session. Again,
this isn't rocket science stuff. 64
bit linux is a big thing for some
people, nice for most and harmless to
most of the rest. Only a few with
custom apps or low loads won't get
anything from it. Unless you *know*
you don't want 64 bit, you want 64
bit. And just because it isn't being
rolled out in bulk doesn't mean it
isn't the thing to do. Quite the
opposite, that means it is the thing
that is being done right now as we
speak and doing a gradual rollout is
the smart way to go in most shops.
\_ I was disputing the "most
places are doing 64 bit" statement
above. Most places are not
doing 64 bit for production. You
allude to that above with your
gradual rollout statement. 32
bit is still far more pervasive
than 64 bit at this time.
\_ If they've got 64 bit in
production right now as part of
a long term roll out then they
are doing 64 bit in production.
Just because they didn't flip
1000+ machines in a day doesn't
mean they're not doing 64 bit.
They're being smart. Any place
with that many machines is going
to do almost any change like
this in a gradual rollout. Why
is 'production' so hard to
grasp? What is your tech focus?
Programming? Sysadmin? Manage-
ment? Something else?
\_ We are 3/4 the way to converting to 64-bit everywhere. We should
be done by end of year. -Ops guy at company with 1000+ servers
\_ Why are you converting?
\_ Because the 32 bit limits suck for real computing. !gp
\_ Then you should've been running Solaris all along. Why
was 32 bit okay 12 months ago and suddenly not good
enough now? I can think of some reasons why you'd have
to upgrade to a 64 bit OS on a 32 bit processor, but they
are not common ones - mostly a need to address massive
files or memory - hardly a reason to upgrade every
server at a company for the hell of it.
\_ Solaris = not free. 32 bit 12 months ago = sucked then
with lots of work arounds. 64 bit now = it just works.
Convert everything because it is easier to maintain
fewer images/builds. This isn't rocket science. Why
do you think a place with 1000+ servers doesn't have
'a need to address massive files or memory'? 2 gigs
is hardly 'massive'. I've got games that require more
ram than that..
\_ Solaris is free. You can have more than 2GB of
RAM or a 2GB file with a 32 bit OS. You don't
need a 64 bit OS for that.
\_ Solaris is not free when you have to buy real
hardware to run it and you knew that. If you're
talking the x86 version, get off the motd.
\_ I didn't know Linux came with free hardware.
\_ The OS is now free, x86 or not. The hardware
is not free, but neither is Intel hardware. If
you are worried about x86, what does Linux
64 bit buy you?
\_ The OS has been 'free' for years but
useless without the hardware. And Sun
hardware is way more expensive than x86.
If I want a production quality system
running Solaris it will cost more than
a linux 64 bit system. That is why Sun
is dead and linux continues to grow.
\_ You can run Solaris on x86 and the
hardware costs will be the same. If
\_ Solaris x86? Whatever. We are
talking about production 24/7 systems
that real people support, not your
Quake4 server.
\_ The code is the same. I repeat,
this is not x86 circa 1992.
\_ Quake4 server. 2007.
you want the Sun hardware then the
premium is not as bad as it used to
be. An Ultra45 costs maybe $6K
versus almost $5K for a high-end Dell.
\_ Great, I just spent an extra $1k for
what exactly? And how much does Sun
support cost on that box?
\_ For a true 64 bit CPU.
A Linux 64 bit system running on a
32 bit chip is not true 64. If you
\_ Who said these are 32 bit chips?
Where'd you get that idea from?
\_ Which 64 bit chip are you
running on then?
\_ So you're going to claim that
the current gen Intel/AMD x86
chips aren't 64 bit? This is
going to turn into a philosophy
debate on instructions sets now,
huh?
\_ Yes, I would claim that
the current gen chips
are not true 64 bit chips.
They are 32 bit chips
with 64 bit extensions.
\_ distinction being? as in,
why should I care?
want 64 bit Linux you need to run on
something like Itanium and that's
not cheap either. However, did you
\_ Who said Itanium? Did we flash back
to 1998? This is 2007. Both AMD and
Intel are selling true 64 bit chips.
Hello?
\_ Yes, and Intel's is called the
Itanium - IA64.
\_ See above.
realize that Solaris x86 is not the
same crappy product it was 15 years ago?
\_ Sun is hardly dead. You can get a 16
\_ Solaris x86 is the same crappy
product it always was. It is a niche
product which makes no sense in 99%
of the real world.
\_ Solaris has a lot of features
that Linux does not have.
\_ Solaris does but who is running
Solaris x86 in a real production
environment? I'd be surprised
to hear of any place with more
than a handful in 24x7 and have
a heart attack plus a stroke if
anyone is doing thousands of
solaris x86 anywhere.
\_ Why does Solaris x86
bother you so much? It's
the same OS as Solaris
SPARC. A lot of Canadian
companies are using it,
FWIW. I would consider
using it in order to use
some Solaris features
like ZFS (until Linux
gets it), the scheduler,
and containers.
\_ Sun is hardly dead. You can get a 8
core T1000 for $5k these days and its
operating costs will be less than 1/4th
what 4 dual core Intel boxes would be.
\_ What about 1 dual chip quad box? :)
The T1000 is not so useful for
floating point, but it was good
to mention it. I had almost
forgotten, since I mostly care
about floating point. Also, IBM's
new Power6 chip will run Linux,
too. To claim you need 64 bit
computing and then run it on a 32 bit
CPU is laughable. I will
eventually run 64 bit Linux on
i386, too, but to think that most
people (or even a sizeable fraction)
are doing so now is deluded. We
are just now starting to port over
most of our major s/w, although
we started a year ago, and it
will probably be another two or
three years before we can drop 32 bit
entirely because of all the testing.
A rule of thumb is that it takes
about 5 years to throw old
hardware/software out the door,
so I expect there will be a lot
of 32 bit Linux for some time
yet. Realize that there are companies
still running VMS and DOS.
\_ Our older 32-bit servers needed to be upgraded, so we decided
to roll out new 64-bit servers with 16GB of RAM. We have lots
of apps that need to address more than 2GB of RAM, or we can
make better use of the new faster boxes by addressing more
RAM in the java container. It is easier in the long run to
only have to support one platform, rather than two, also. |
| 2007/7/6 [Computer/SW/OS/Solaris] UID:47193 Activity:nil |
7/6 Hi I need some general pointers of what to look for if my
app is running slow on a recent Solaris box. What should
I be noting? |
| 2007/2/12-15 [Computer/SW/OS/Solaris] UID:45722 Activity:nil 50%like:45719 |
2/12 Trivial Solaris 10 and 11 0-day exploit. AWESOME!
http://isc.sans.org/diary.html?storyid=2220
-dans
\_ That's why I run PC-Dos. It's never been remotely hacked. |
| 2007/2/12 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:45719 Activity:nil 75%like:45716 50%like:45722 |
2/12 Trivial remote root exploit in Solaris 10 and 11. AWESOME!
http://isc.sans.org/diary.html?storyid=2220 -dans |
| 2006/9/10-12 [Computer/SW/OS/Solaris, Computer/SW/Database] UID:44336 Activity:nil |
9/10 Has anyone compiled MySQL on Solaris 2.8 and had problems when
it gets to mysql_tzinfo_to_sql? I'm getting:
ld: fatal: library -lpthread: not found
ld: fatal: library -lthread: not found
ld: fatal: library -lposix4: not found
and make is failing. I'm compiling MySQL 5.0.24, 32 bit.
gcc 2.95.2. I just tried it --with-mit-threads, but that didn't
work either. Please email me. I want to burn the entire server
room at this point. Thank you! -sax
\_ I never had your specific problem but it looks like either you
don't have those libraries installed or your library path is
broken.
\_ I have built 3.23.xx and 4.0.x on sol 8/sparc with both the suncc
and gcc. Have you tried the flags MySQL uses?
http://dev.mysql.com/doc/refman/5.0/en/mysql-binaries.html
\_ Thank you both! There were a few more command line options
I was missing that helped. I was using
--with-mysqld-ldflags=-all-static, which apparently doesn't
work on Solaris. My biggest problem, though, was that it
looks like this version of gcc was installed incorrectly.
Switching over to gcc-3.2.1 solved the rest of my problems.
-sax |
| 2006/6/22-28 [Computer/SW/OS/Linux, Computer/SW/OS/Solaris] UID:43474 Activity:nil |
6/22 Anyone here deploy Linux-based Sun Rays lately? Thin clients
sucked a few years ago, but Sun claims that performance is much
better now. If performance is decent, I'm interested. Lots of
people at work just use their desktops as terminals anyway.
\_ we have a few Sun Ray clients here. They have pretty much just
worked since we set them up.
\_ How is performance?
\_ Fine for clean X apps. Sun's Java Desktop Suite is a dog
regardless of what you run it on. Note that we don't use
them for Linux desktops yet.
\_ i do that all the time for POC purposes. I prefer *LINUX*
over Solaris, against my company's party line, simply because
Linux is still a better desktop OS than Solaris. What kind of
information do you want to know? kngharv
\_ What makes a desktop 'better'? |
| 2006/4/25-27 [Computer/SW/OS/Solaris] UID:42822 Activity:nil |
4/24 http://news.bbc.co.uk/2/hi/business/4941050.stm Sun Microsystem's McNealy quits post as head, making it a bright day for Sun shareholders, with price rising to 6% to $5.29 in after-hour trade. \_ Big baby finally grows up. |
| 2006/4/3-4 [Computer/SW/OS/VM, Computer/SW/OS/Solaris] UID:42629 Activity:low |
4/4 I'm setting up a simulated infrastructure for a client using
a bunch of vmware sessions. Originally we had these on different
boxes, but I only have one machine available. As there individual
images require different subnets, can anyone think of an easy way
to have vmware sessions on, say, 3 different subnets exist on the
same physical machine and talk to each other via some sort of
"virtual router"? -John
\_ Add a virtual interface for each subnet on the vmware host box
and have it route.
\_ This sounds like it would work. I have considerable experience
running many virtual interfaces on Linux and Solaris, email me
if you want info on how to do it. -dans
\_ it is hard to create virtual interfaces?
\_ No, it's a couple line changes in a config file (which one
varies by Linux distro or Solaris) or a manual ifconfig
invocation. Of course, figuring out what config file and
what options takes a while if you don't know what you're
looking for and which man pages. -dans |
| 2006/2/22-27 [Computer/HW/CPU, Computer/SW/OS/Solaris] UID:41954 Activity:nil |
2/22 Can someone give me a link to a ISO of opensolaris
that I can burn and boot on my PC? thanks
\_ Opensolaris distributions are still pretty much for developers only.
If you want Solaris for your PC, download Solaris 10/x86 from Sun
ISO(s) from Sun.
\_ The closest thing to an official ISO for openSolaris is
sxcr. You have to register w/ sun to d/l it:
link:tinyurl.com/nswzr (javashoplm.sun.com)
One of my friends on the openSolaris team suggested that
you try nexenta instead |
| 2006/1/2-4 [Computer/SW/WWW/Server, Computer/SW/OS/Solaris] UID:41196 Activity:nil |
12/3 Hello, I'd like to setup a wiki and a discussion board for
people interested in a particular niche market I'm looking at
(I can't give out details because someone may steal it). I
already colo my family web site (<lastname>family.com) with a
friend of mine on Solaris at InReach, Oakland. What's the best
software to get to host a wiki and a discussion board, and do
I have to setup suexec and mysql? |
| 2005/11/22-23 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40690 Activity:nil |
11/22 Is there a native process in Unix (SunOS 5.7) that logs port traffic?
Barring that, do you know of any good software that would do the trick?
SdTFW for "logging port traffic unix" but not finding what I'm looking
for. TIA. --erikred
\_ tcpdump -w
\_ snoop. Ethereal. -John
\_ snort?
\_ You want something solaris native, 'snoop' is it.
\_ Once again, motd >> Google (and my google skills). Thank you.
--erikred |
| 2005/11/14-15 [Computer/SW/OS/Solaris] UID:40576 Activity:moderate |
11/14 Ugg. It took me all of 20 seconds to figure out NIS on linux.
I can't even figure out how to lock a yppasswd on solaris. Help.
\_ Hint: Your issue is not with NIS
\_ Uhh, what is it with?
\_ Solaris' passwd mechanism doesn't provide a locking mechanism
(unless it's a recent addition i don't know about)
\_ from shadow(4):
The lock string is defined as *LK* in the
first four characters of the password field.
--Jon
\_ further note, you can use NP as the passwd string
which will allow uid to run cron jobs/etc --Jon
\_ Which you should not be using anyway...
\_ Well, I don't fault people using NIS in a secured corporate
environment.
\_ Hard on the outside, soft and chewy in the center.
\_ I've worked on a lot of networks and they have all
been like this (well, hardER on the outside anyway,
some were pretty soft all around).
\_ Probably, but with NIS you may as well not even
bother with security.
\_ NIS really isn't that bad in terms of
security if you have strong passwords
enforced. NFS, that's another thing. -tom
\- you need more than strong passwds. you need
tight securenets, you need to not let
people log into servers, you probably need
the servers hardcoded to the clients etc.
in many environments sniffed credientials
are now a bigger problems than cracked
passwds. once somebody gets unauthorized
access to an unpriv nis account, it is
highly likely they will be able to find
some local exploit. and without the other
issues raised above you are potentially vuln
if a machine not even in your domain but
just within your bcast domain is rooted.
that being said, i think nis has its place
but that is beyond the scope here.
\_ I didn't build the system, I was just hired to make
a few changes, not rebuild it. -top
\_ One good and fairly secure alternative to NIS, if you don't
want to go with with LDAP, is to setup cfengine to rebuild
/etc/{passwd,shadow} files on all machines. The downside of
doing this is that if someone roots a client box they can
still see your local /etc/shadow file. This sort of thing
could be prevented with LDAP. |
| 2005/11/1-3 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40376 Activity:nil |
11/1 what's the name of a general login machine in berkeley cs?
please email dpetrou@cs.cmu.edu. -dpetrou
\_ argus
\_ http://inst.eecs.berkeley.edu/cgi-bin/clients.cgi
That is a list of everything. -mrauser
\_ failure to read directions. |
| 2005/10/28-31 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40320 Activity:moderate |
10/28 Docs say I can't do dd from disk that is in use.
This seems (based on my bad memory) to be contrary to my experience.
Any opionions on duping an inuse disk (to a non-inuse one).
this is on solaris 8.
\_ live update, ufsdump, rsync, dd... you want your filesystem
as quiesced as possible though for all of the above
\_ what the above and others say about quiet drives and RO and then:
since you don't tell us what problem you're trying to solve, my
very generic advice is this: dd the drive so you get boot blocks
and other fun stuff you might want, then do an rsync after that just
for kicks in the hopes of catching a few of the files that might
have been in use while doing dd. no, this followup rsync will not
make a perfect mirror on an active disk but if you wanted to do a
real backup, you'd use real software for it or at least take the
source drive offline. you're well into kludge territory already,
anyway.
\_ You can dd from a disk that is in use. You will obviously not
get an exact duplicate and, as above, I would not try this
on a very busy filesystem. However, it works fine under Solaris
9/10 and Red Hat Linux. I've done this to make 'backup' disks
(successfully) many times.
\_ If you need to keep the filesystem mounted RW, I'd suggest
using rsync rather than dd. If the filesystem is mounted RO
it seems like dd would be fine.
\_ Rsync is slower and won't grab the boot blocks (if you want
them).
\_ rsync is NOT slower than DD unless the filesystems are
basically full. -top
\_ Yes, it is. Best case scenario (using the right
rsync options) they are about equal. "dd" can use
the raw device and bypass the filesystem entirely.
Amended: You are right that if the filesystem/disk is
not very full then "dd" will be slower merely because it
is copying "empty" space but not because "dd" is
intrinsically slower.
\_ What problem are you trying to solve?
What are your priorities?
\_ so it's not "intrinsically slower", it just takes
longer to run? Get a clue. -tom
\_ I think you need the clue, tom. "dd" is
faster. If you want to prune the list of files
you want to "dd" first it will kick the butt
of "rsync". You know that, too, but you just
like to argue. When copying the exact same
volume of data, "dd" is faster.
\_ While driving a car is faster than walking,
walking from Telegraph and Bancroft to
Telegraph and Durant is much faster than
driving. "dd is faster" is meaningless
unless the problem is well-defined. -tom
\_ Even your example is false.
\_ Alright, we're convinced: you're an idiot.
You don't have to try this hard just for
us.
\_ You may as well say that a plane
is slower than walking, too. The
statement about a 'well-defined'
problem is your own parameter but
doesn't change the fact that
planes are faster than walking.
\_ To put this in a way that even
an idiotic computer geek might
understand, if you have two
different sort algorithms,
one that's usually faster in
practice but O(N^2), and one
that's usually slower in practice
but O(N log N), it makes no sense
to say that one is intrinsically
faster than the other. It
depends on the problem you have
to solve. -tom
\_ HAMMER GOOD!
If you took a second to think, you'd realize
you're clueless. Or maybe not. It's dim.
\_ Do you know what the word 'intrinsic'
means?
\_ Yes. You seem to have trouble with the
word 'slower', though.
\_ "dd" is intrinsically faster.
\_ Not for the problem posed (ergo
"HAMMER GOOD"). If you want to
keep spouting your technically
correct and completely irrelevent
point, by all means... You will
still be wrong.
\_ You have come to know that which
is tom.
\_ hey moron: he's replying to
!tom.
\_ We don't even *know* what the
problem posed is yet. However,
we do know the guy wants to
use 'dd' to solve it. Rsync
may not even be a viable
solution in this particular case.
\_ What problem are you trying to solve? What are your priorities?
\_ try using solaris' Volume Manager (also known as disksuite).
create a one way mirror containing the current live FS as the
initial submirror, then attach the second empty FS as another
submirror. wait for the second submirror to sync up, then at
your leisure, detach the second submirror.
look at metadb, metainit, metattach, metadetach, metaclear,
and other "meta"commands. --Jon
look at metadb, metainit, metattach, metaoffline/online,
metadetach, metaclear, and other "meta"commands. --Jon
\- i have a solaris 10 sunblade 1500 which some colleages
wanted to clone to save time. i rebooted my machine and used
/usr/bin/dd if=/dev/dsk/c0t0d0s2 of=/dev/dsk/c0t1d0s2 bs=1000k
and thne fscked. all the machines work fine ( 5 clones). it
sounds like you know what the issues are so just take this an
a data point. 73gig disks ... i think it tok 40min or so.
oh i am sure i booted it su and killed off most of the
processes before running dd.
\_ The thing with the Solaris LVM method is that you don't
really need to have the filesystem quiescent. You get a
valid, consistent snapshot at the time of metaoffline or
metadetach. --Jon
\_ As long as you understand this isn't considered "production
quality", if that worked for you, great. I wouldn't do that
for something going into a data center 24x7, etc.
\- i bet lot of production operations were running on
less reliable windows installations (i dont know how
reliable windows OS is today). anyway, these are
cheep sun workstations with ide disks, so what do you
think. anyway, it seemed to be the OP was aware of the
"issues" and was just interested if it was "crazy talk"
[you'ld be lucky if it worked] vs. 99%chance it should
be fine. hence the "datapoint x5" comment.
\_ who said that just because you put a windows box in
production means its production quality? thanks for
pointing out they're workstations. that was so unclear.
i was pointing out that this isn't production quality.
sheesh.
\- are you a frequent visitor to Casino Troll?
\_ And what is "production quality"? This is essentially
what tools like "System Imager" do. What are you
afraid of? 'cksum' the OS if you want to. Bits
is bits.
\_ PQ: built correctly, not a hack job with dd
from another box. Something you'd bet your
career on. Workstation for the guys? Whatever.
Server in the colo running critical db? No way
is that getting built from a dd clone off a hot
drive. Building a few dozen/hundred/whatever
web/whatever servers? Jump/Kick Start. I've
done the clone thing for workstations. I'd
never do it for servers. If you want to discuss
it in detail, post your name and we'll take it
to email. |
| 2005/10/26-28 [Computer/SW/OS/Solaris] UID:40281 Activity:low |
10/26 Any recommendations for Project management software, must be
free$$ and run on Solaris/Linux/Windows. dotproject?
\_ Hate to say it, but MS Project is the de facto standard. It
is very expensive. I haven't found anything free that does
close to what it does.
\_ Which is what? I looked at it in the past but wasn't impressed,
though it's likely I didn't know what to look for. --darin
\_ Help manage a project? Scheduling, milestones, resources,
task management and planning, Gantt charts, etc. |
| 2005/10/16-18 [Industry/Jobs, Computer/SW/OS/Solaris] UID:40129 Activity:nil |
10/16 So any sun old timers still reading the motd? I was recently named a
SARC "intern" and I wanted to find out if anyone has been through the
"intern" process and if they had any tips/pointers. txn.
\_ I'm a Sun "old timer" in several ways, especially if you count me
playing with the first prototypes when I was 11 years old ... I
was a Sun intern in summer of 1991 and had a blast. My take from
that was that if you actually do useful work and get anything
accomplished you can get a job there once you graduate (I did).
You get to meet Scott McNealy and hear his puppy calendar stories
and for me it was the most fun I've ever had in a job, especially
with the huge water battle between SunSoft and SunLabs. -eric
\_ maybe I wasn't clear - I'm a staff eng at sun and I was put
on the Software ARC (arch. review committee), and I wanted
to get some pointers on what ARC members do, &c.
\_ Nope that wasn't clear. I joined Sun as an MTS-1 and only
worked there 2 1/2 years so I never got anywhere in that
regard -eric
\_ neato. sounds like a paper pushing job.
\_ From what I can tell it is mostly paper pushing, but
this is in addition to my regular job, so it just
means more work for me. |
| 2005/10/13-14 [Computer/SW/OS/Solaris] UID:40077 Activity:high |
10/13 I know how to set up nis master/slave services on a solaris (8)
box. But given already set up services, how can I tell which
is master and which is slave?
\_ 'ypwhich -m' is what you want. The master server is encoded into
the NIS maps themselves. -ERic
\_ ypwhich? iirc, the master should tell you that it is its
own master and the slave should tell you that the other
machine is the master. I'm not sure if this works for nis+.
\_ hmm, really? in linux-land i have seen SlaveNIServerX
respond to a yphich which "SlaveNIServerX"
\_ I think I'm wrong, maybe 'ypwhich -m hosts' (or some
other map) might work.
\- you can use something like this:
rpcinfo -p <host> |grep -q ypxfrd && echo "ypmaster" || echo "not_ypmaster"
rpcinfo -p <host> |grep -q ypxfrd && echo YPMASTER || echo NOT_YPMASTER
obviously you can loop over something like
foreach i (`ypcat ypservers`). --psb |
| 2005/10/7-9 [Computer/SW/OS/Solaris] UID:40017 Activity:nil |
10/7 I'm using winxp's rsh to connect to a solaris 2.8 machine. Is there
anyway to have the rsh connection timeout? Currently, it seems to
hang forever when something goes wrong. So far the only remedy is
to reboot the windoze machine. TIA.
\_ I don't know anything about winxp's rsh but generically speaking
you can wrap anything with perl or expect alarms/timers.
\_ I have some dos scripts which invoke certain programs on
solaris. Sometimes these scripts are invoked from other
scripts and/or programs, but when someone uses ctrl-c, the
rsh connection isn't closed properly, so the solaris host
will refuse further rsh connections for the user in question.
So on windows, rsh is no longer running (you can't kill it),
but if you reboot the windows host, it will work again. Using
netstat on solaris shows that it's waiting to close for the
windows machine even though rsh on windows is no longer running.
\_ use cygwin's rsh then.
\_ would this rsh close the network connection in the above
scenario? our build machine doesn't have cygwin installed
yet since our build-meister is still debating on whether
to include it or not on std build machines. |
| 2005/7/25-27 [Computer/SW/OS/Solaris] UID:38821 Activity:kinda low |
7/25 I have a MacOS 10.3 box that cannot connect via ssh to Solaris 9
and Solaris 10 boxes. MacOS 10.2 and MacOS 10.4 work fine.
Connections to Solaris 8 and Linux (from 10.3) work fine. What's
up with 10.3 such that it won't work with Solaris 9 and 10?
\- i ssh from my 10.3 powerbook to s9 and s10 many times a day
for many months. this is to sun and openssh daemons. so your
problem is local config and not some inherent interop issue
probably.
\_ Can you supply which SSH version numbers you are using? It is
not confined to any one Mac system or any one Solaris
system. I can try many different Macs to many different Suns
and have the same problem, but never with Solaris 8 and
never with MacOS 10.2 or 10.4. Error is always: "Operation
timed out." Nothing logged. Tcpdump shows lots of acks with
no responses.
\- if you just do a "telnet s9machine 22" do you get a
version banner?
\_ No.
\_ I think you are running into the infamous ipv6 dns issue w/ 10.3.
One way to fix this is add the same nameserver as the primary
and secondary nameserver.
\_ I did this and also turned off IPv6 in /etc/hostconfig. No joy.
\_ Okay, that is VERY strange. Are all the boxes on the same
subnet? Maybe this is a routing issue.
Can you ping/traceroute the S9/10 boxes from 10.3?
\_ Various subnets and I cannot ping or connect to,
say, port 25 from MacOS 10.3. I *can* do these things
from other OS, including MacOS 10.2 and MacOS 10.4.
However, to Solaris 8 from 10.3 is fine. This started
after upgrading to 10.3 from 10.2. So far all systems
I have tested (3) do this. They worked fine before the
upgrade (fresh install, actually) and are on 3
different subnets.
\_ Firewall or tcpwrappers? -ax
\_ That are selective to MacOS 10.3? |
| 2005/6/29-30 [Computer/SW/OS/Solaris] UID:38352 Activity:nil |
6/28 Solaris 9/10 local root exploit:
http://archives.neohapsis.com/archives/fulldisclosure/2005-06/0344.html |
| 2005/6/23-25 [Computer/SW/OS/Solaris] UID:38267 Activity:nil |
6/23 Stoopid Tech Q: Is there a version of CDE for SunOS 5.7?
\_ Updated Q: how the hell do I get the CDE running from the
command line?
\_ /usr/dt/bin/dtsession?
\_ got it, thanks.
\_ turns out /usr/dt/bin/dtlogin -daemon does the trick.
\_ Oh. in that case, /etc/init.d/dtlogin start |
| 2005/6/23-25 [Computer/SW/OS/Solaris] UID:38262 Activity:nil |
6/23 Help! I am trying to convince my collegues that Solairs 10 is not
even close to be used as a desktop OS. They've kept telling me that
since it uses Gnome UI, it has the same usability as Linux. How
do I counter that? So far, I can say:
1. Linux has a lot more working device drivers
2. Linux has a lot more consumer-oriented, multi-media applications
anything else? sun guy
\_ Your counter to Solaris is Linux? Grow up.
\_ that is all the choice I have... lesser of two evils
\_ I suppose you're talking about Solaris on intel?
1. Is not very relevant if you intend to use Sun hardware and
you don't mind paying their prices (it's not like Sun's
opteron boxes are overpriced, but they're designed to be
'workstation' class boxes. Don't even think of getting one
for like $1000 or less unless you're an education customer).
2. You could make a good case if you actually have such a software
package in mind (e.g. I wouldn't have a hard time convincing my
users to stay away from Solaris on intel just because Matlab or
some other packages they use don't work on it)
\_ I think Linux makes a really nice desktop. Maybe Solaris does,
too. I haven't had it on the desktop since Solaris 8 and I
don't miss it. Personally, I like Windows (in a Linux VM) or
MacOS as my desktop. |
| 2005/6/14-17 [Computer/SW/OS/Solaris] UID:38129 Activity:nil |
6/14 open-sourced version of Solaris went on-line.
http://www.opensolaris.org
I tried to get the message that at least they should
LGPLed device drivers... the message never reach the top...
Sun Guy
\_ I tried to get the message that they should have
released it w/ a BSD license and included java
along w/ it, but that didn't happen either. - yaSunGuy
\_ shoot me an email... we talk about this off-line.
kngharv
\_ I tried to tell Sun that Linux/Intel would really hurt them and
their sales guys laughed and said that the server market is
what they wanted anyway because it is where the profit is. I
wonder how that's working out for them. |
| 2005/5/5-6 [Computer/SW/OS/Solaris, Computer/HW/Laptop, Computer/HW/CPU] UID:37529 Activity:kinda low |
5/5 I am not familar with Solaris, so, I may be biased... My impression,
from a desktop/laptop user point of view, is that Linux has
better userability than Solaris10, eventhough Solaris 10 also come
with a gnome desktop environment. However, I find myself in a
rare position of not knowing how to articulate this point to
some of the Solaris fanatics. Their argumenet is that since S10
is also equipped with GNOME, userability is the same as Linux.
1. is this arguement flawed?
2. if you think Linux indeed has better usability, any concret example
of userability is more than gnome desktop TIA
\_ From a user point of view, the biggest advantage of Linux is
all the software it ships with. All of that *can* be added to
Solaris, but it isn't there out of the box. Another advantage
of Linux is that for most tasks the actual hardware is faster,
assuming you are running Intel/AMD. Sun has the advantage for
problems that fit within the larger CPU cache, but are larger
then the (smaller) Intel cache.
\_ No Linux desktop user has any basis for say his desktop is superior
in usability to the S10 desktop. This is like the dateless ugly
girl at the dance picking on the other ugly girl for not having a
date.
\_ Thanks for erasing my post and adding bullshit at the same
time.
\_ it's ok, I read it already. -- OP
\_ that is why I post this on motd. If i think i am in a
position of being objective, i probably can make assessment
on my own. having said that, from little things such as
ls, man command, to the way driver installed on S10, I
still think Linux is friendlier.
\_ BWWAAAAHAHAHAHA! Man, it's 'friendlier' until you have to do
low down systems level programming. Then it's 'friendly'
in the same way your 320lb cellmate 'Butch' is friendly.
\_ ^320lb cellmate^neurotic 320lb cellmate
\_ Is that what the server world has really come down to? One
version of 'ls' is "friendlier" (meaning: the way you learned
it on the one system) therefore that system is superior? I'm
curious how many non-Linux systems you've used and for how
many years each.
\_ Your point is correct, but I've just watched (again)
a bunch of presentations tearing gaping new assholes in
Windows security. Maybe he should have mentioned this.
Usability is actually pretty decent. -John
\_ What about drivers? Unless you buy hardware from Sun, you're
going to have some serious problems with hardware support. In
addition, the number of binary-only desktop applications seems to be
dwindling (for example our users would like to be able to use the
latest Matlab and Acrobat reader but those aren't available on
Solaris/x86). In addition, even though Sun includes Gnome, I doubt
Solaris x86 comes with as many popular open source applications as
there are in a typical Linux dist (but then I could be wrong, I
haven't used Sol10 myself yet although Solaris 9 loses hands down in
this area even though it also comes with Gnome).
\_ Easy fix for that: just buy Sun hardware.
\_ I don't understand the desire on the part of some people to attempt
to compare the quality/friendliness/whatever of the desktop
environments of what are really server systems. If you want a nice
GUI, buy a Mac. If you want to play games, work in an office, and
have the largest variety of desktop apps, buy Windows. If you want
a server system and you're the one making the buying deicision, the
question is not "Is this Gnome better than that Gnome?" The
question you're answering and the 'debate' you've involved in is
meaningless.
\_ Someone has to develop the apps for the server system and
that lucky person often gets to use it as a desktop. It is
true, though, that using a PC/Mac as one's desktop and
remotely connecting to the server to code makes sense
depending on the app being developed.
\_ Yep, I've got the kind of luck where I have to use CDE
at work. As far as desktops go, Linux is easier to use
on x86 because there are just a lot more people using it
and improving it. Also, I'd have to say GNU utilities are
generally superior to default solaris ones. |
| 2005/4/29-5/1 [Computer/SW/Security, Computer/SW/OS/Solaris] UID:37425 Activity:moderate |
4/29 Why did Sun decide to implement SMF in Solaris 10? Was it just to
piss off customers or is there some technical advantage?
\_ Are you talking about the new way to start/stop programs,
&c.? If so, I must agree that the only purpose was to piss
off customers and prove that Sun can do something stupid and
different than Linux (chkconfig may not be great, but it
mostly works and everyone knows how to write init scripts)
BTW, SMF pissed off a lot of ppl inside sun who have to ship
products on other *nix than Solaris.
\_ I guess Sun should be on http://fuckedcompany.com if it isn't
already.
\_ Can someone give me a list of reasons why SMF is bad?
\_ Complicated new way to do something that has already been
done. Like I said, if there's some technical advantage then
I'd like to know what it is. Maybe there is one. If not, it
is just change for change's sake.
\_ 1. SMF uses non-standard commands - you can't simply
start/stop a process by calling its init script,
you have to know what its SMF "name" is. Even if
you don't have to deal w/ other *nix, SMF makes
switching btwn S9 and S10 a pain.
2. SMF enable/disable semantics are bizarre - you
can't just say enable/disable X like in chkconfig
and assume that the daemon is enabled
3. SMF fails to provide adequate feedback re failures
of configuration. Often, you can't tell if a fault
needs to be cleared in order for it be enabled.
service can be enabled.
4. SMF's files are non-standard and their contents
are not explained well - the purpose of SMF is
to make fault recover/mgmt easier, however if
most of your admins don't/can't figure out how
to fix config problems, faults will take longer
to remedy. Developers and Admins should not have
to read some guys blog on http://blogs.sun.com in order
to get details on how the system works.
\_ http://docs.sun.com/app/docs/doc/817-1985/6mhm8o5n0?q=smf&a=view
\_ Yes there are docs, but the docs don't
really have anything useful in the. Ex.
tell me where in that page it specifies
how to use svcadm to disable a process
from being started or how to tell if
the reason a particular process is not
starting automatically is b/c SMF thinks
that the process is in 'fault' state and
must be cleared.
5. The fault mgmt functionality provided by SMF could
easily have been provided through additions to
existing functionality (specific args to init
scripts, allowing apps to dump monitoring scripts
into a given directory, &c.)
\_ Sounds like one of those numerous cases where Sun was trying
to solve the problem which has been already solved
by others and comes up with some terribly complicated and
non-standard way of doing things. *sigh* |
| 2005/3/9 [Computer/SW/Unix, Computer/SW/OS/Solaris] UID:36604 Activity:low |
3/9 After I start an xterm in SunOS 5 and Linux, is there any command to
change the title of the xterm window? Thanks.
\_ Check ~geordan/bin/retitle
\_ That 2 should probably be a 0 -- that way the new title stays
even when the window is minimized (or whatever the standard
X term for that is). -alexf
\_ using 1 will change the icon title. using 0 changes both.
\_ I always just type:
% cat
^[]0;New Title^G
^D
(^[ = escape, ^G = ctrl-G, ^D = ctrl-D) --dbushong |
| 2005/2/23-24 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:36382 Activity:nil |
2/23 Trying to load tcsh on a SunOS 5.7 machine, getting C compiling
errors. Resources? Suggestions? STFW returns nothing useful.
\_ what errors?
\_ Trying to load it or compile it? If you are just trying to load
it, get it from sunfreeware and pkgadd it. |
| 2005/1/5-7 [Computer/SW/OS/Linux, Computer/SW/OS/Solaris] UID:35563 Activity:very high |
1/5 So who else thinks that Linux Kernel Development has gone haywire?
WTF is up with this movement from an 8k to a 4k stack in the kernel
that breaks tons of existing drivers that are ported over from
Windows? And wtf is this crap doing on production distros like
Fedora? Don't they realize that if you're going to have a large
install base that you can't arbitrarily do crap like that anymore?
\_ I agree that they are lame and have always have been, but
Fedora isn't a production distro. That's RHEL.
\_ So in other words RH just became even dumber than they used to
be by foisting Fedora on the user community and charging
for the bugfree version.
\_ Fedora is a development platform; that's how it is positioned.
If you don't want a development platform, run RHEL, or
debian or whatever. It's not being "foisted" on you. -tom
\_ No shit sherlock. But the problem is that usually what
happens in Fedora is just reflected in RHEL. RH being
the dumbass company that it is obviously doesn't
do anything like do a real-world usability test on
its distro so going from one major release to the next
results in all your binaries being broken. Also,
a lot of end-user end up using Fedora because they
stopped distributing RH, so in effect it is being
foisted on the userbase with the said userbase
complaining about things being broken.
\_ you're a moron. -tom (really)
\_ you're tom. -idiot
\_ Ouch, now THERE'S a harsh insult.
\_ I'm not sure what part you are objecting
to, but RH's pricing structure for EL has
driven lots of people to use Fedora Core
as a production OS. Many times it is hard
to justify the added cost of installing
EL and a customer choses to deploy FC.
You don't really have a choice but to
support FC as a application developer.
It isn't really practical to tell a customer
to install Debian 3.0R3 or something.
\_ yeah but the bug free version is GPL also. you can try to
use CentOS or one of the other RHEL redistributions.
unfortunately they still suck as a consumer OS.
actually <DEAD>scientificlinux.org<DEAD> looks interesting.
\_ They break drivers all the time anyway as far as I could tell.
You're supposed to stick with some old kernel for a long time for
actual consumer use. But why would you need drivers for Linux?
It's not like you can play games or really do anything anyway.
\_ Well, unfortunately since Sun did such a bad job maintaining
market share us EDA folks are being forced into Linux. Now
we have to do do crap like recompile the kernel just so the
stupid display driver works.
\_ Yeah I use that stuff at work. As long as other people are
responsible for making it all work I don't really care.
\_ I compare the adoption of linux by corporate america to the
ubiquity of windows. Some mid-level managers and idiot salespeople
who thought it gave them cache foisted it upon the world where it
went batshit crazy and drove us all insane.
\_ I actually prefer Linux to, say, Solaris or HP-UX. It has
its limitations, but overall it is cheaper, faster, and
easier to maintain in many ways.
\_ ditto. -- SUN guy
\_ No offense, but Solaris is a far better
operating system. Just because for a long
time Solaris didnt ship with perl and you
have to build you own tcpdump doesnt make
it otherwise. If you get involved in the
innards of operating systems, this is pretty
clear. There are some SysV things that
arent ideal, but if you are trying to debug
low-level things, it is pretty clear.
\_ Not to mention that drivers actually work
in Solaris...
\_ Linux has far more working drivers than Solaris.
Solaris just works on the very limited hardware
Sun provides. -tom
\_ I work for SUN and I've been fighting on driver
issue everyday. And I can tell you flat out
that you may think driver works on Solaris, but
Linux is the only way to go. People would
write Linux drivers, but SUN relys on 150 people
in Beijing to crank out those things one by
one. As hard as those Chinese monkey works,
they can never match the speed which hardware
comes up.
\_ You must live in some other universe. I work
for Sun and we have the hardest time getting
drivers to work for even simple stuff like
gigE nics (ex E1000 driver on S10 was a
nightmare for a long time). And you can forget
about AGP in most cases. Some big shots felt
AGP was the shits so no support in Solaris.
There were several cluster deals we couldn't
bid on b/c there was no AGP support in Solaris.
\_ hey, would you mind if I contact you?
-another SUN guy (id 152093)
\_ I think you may be missing the forest for the trees
here. How many people spend their time debugging
``low-level things?'' How many people just want
the system to come with a modern version of perl?
Once you reach a critical threshold level of
stability (which Linux hit some time in 1999 or
so) comparing OS internals dick size becomes
pointless.
\- if you want to say linux is more useful
because i can surf my p0rn and play my mpegs
"better" that's fine. useful to me !=
better os design. it's not a matter of
how many people do this. it's more like
looking to a kernel crash dump tells you
a lot about what is under the hood.
\_ In the REAL world, most people write
applications that run on the OS. I
can almost understand that Sun doesn't
want to ship MySQL or PostgreSQL w/
Solaris, but WHY IN PARTHA'S NAME
did they wait till S9U3 to ship wget
in /usr/sfw and S10 to add gcc? I
shouldn't have to go to some website
to download badly packaged freeware.
Every single Linux distro comes with
this stuff pre-installed. Oh yeah,
instead of chkconfig and isc dhcpd
I get svcadm and sun dhcpd which are
complete CRAP.
Linux has its own problems, but one
HUGE advantage of Linux is that you
can tell your customers to get RHEL
3 ES or SuSE Pro, install it in
server config and then install your
software on top of it. The same RPMS
every time, in the same location,
it makes it easy to test, debug and
support. Unlike Solaris where you
have to ship all your 3d party pkgs
you don't have to worry about keeping
up to date with DBI.pm fixes, PostgreSQL
security patches, wget vulnerabilites
&c. The OS vendor takes care of that
so you can concentrate on your app.
\_ and for your information, MS Windows hit
that threshold by year 2000 with Windows 2000.
Despite you may not think that way.
\_
\_ I don't. Solaris + Native Sun HW is definitely a lot
easier to setup and better integrated than Linux. Solaris
x86 on the other hand makes zero sense. Sun HW also used
\_ let me tell you something. The biggest mistake SUN
ever made was terminate its Solaris x86 program back in
2000. Since then, Linux took off.
-SUN guy who is trying to sell Solaris10
everyday.
to be quality, of course since the U-Sparc 5/10 days
this is no longer true. HP-UX is basically dead, has been
since the late 90s. I just think it's really lame that
in the year 2004 I have to recompile the stupid kernel
to get something like UDMA to work. In some ways, Linux
sucks because it's just a rehash of 30 year old tecnology
on cheap commodity hardware. I mean, shouldn't there be
something better than what's essentially just glorified
UNIX? In all the years with Linux I haven't really seen
anything that really was groundbreaking in terms of
kernel dev. I mean, wtf was Torvaldis smoking when he
decided he was too lazy to implement a modular structure
to the kernel, and why hasn't this been corrected in the
15 odd years that Linux has been around?
\_ What Torvalds was smoking when he decided he was too lazy
to implement a modular structure in the kernel:
http://csua.org/u/ale
You may bitch, but history shows him to be correct.
\_ "correct"... Linux has become more modular over time,
and other OSes haven't sacrificed their modular design
at the altar of Linus. What exactly was he "correct"
about? That linux beat minix? Big whoop.
\_ Hah, exactly my point. It's like saying that
the Chinese had stopped charging families for the
bullet they execute prisoners with. Going from
the Americans had stopped genociding people
for human rights, freedom and democracy. Going from
crap to not so crappy isn't exactly innovation.
\_ the bigger picture is not about technical superiority.
and i was hoping you guys notice that when Windows captured
98% of the OS market while argueably it is the worse
major OS on the market right now.
\_ No, the point was that Windows 98 was backwards
compatible with Windows 95 which was backwards
compatible with Windows 3.11, etc. Now Linux
version 2.6.6 isn't even fucking compatible with
Linnux version 2.6.5. That's progress?
\_ Man, this whole thread could be summarised as: OP is upset that
Linux community doesn't care about 3rd party drivers, and many
CSUAers continue to deride Linux for not being enough like
X \in { BSD, Solaris, DomainOS, ... }. |
| 2004/12/1 [Computer/HW/CPU, Computer/SW/OS/Solaris] UID:35146 Activity:high |
12/1 open source sadness, SUN being destroyed by linux..
\_ More like Sun stupidity by refusing to either
A) Make Solaris x86 free until it was too late.
B) Give up on the stupid "java thin client" crap
C) Open Source Solaris and let people actually write
drivers for 3rd party peripherals.
D) Standardize on a PC architecture instead of dragging
on forever with a dying USPARC platform.
\_ Re: C), Why does it need to be open source for people to write
drivers for 3rd party peripherals? XP is not open source, but
people can write drivers.
\_ Obviously you don't understand software/hardware economics.
Unless your OS has a large entrenched user base (such as
Windows) nobody is going to spend money on writing
drivers for it. It's economically unsound, especially
if you have to buy the tools. Solaris didn't even come
with a free cc compiler (SunOS did). As for writing
drivers for XP, it costs in the neighborhood of $10,000
to develop a driver for it (I know, I've been involved
in one). Solaris needs to be open sourced so that OSS
people can port Linux drivers to it (which has been done
before) or write drivers from the ground up, especially
on an X86 platform. Apple's OS X would have faced the same
problem if it had jumped onto the X86 platform.
\_ I am glad. I was a big, big Sun customer who kept telling my
reps that we were going to move to Linux mostly because of cost
since we actually prefer Solaris. Most of us don't care
about the OS price (cheap) or whether it is open source.
It's the Intel hardware that was much cheaper and drove the
decision. My reps had the attitude that we were small potatoes
since we didn't buy lots of E10Ks and that it was fine for
client-side to be Linux as long as they could sell their
expensive servers. Now they are paying the price for not
listening to their customers...
\_ Not just that--I've encountered some pretty serious
incompetence here. They had the chance to take a really huge
services contract away from a major outsourcer--they threw a
half-wit project manager and two untrained sysadmins without
work permits at it, because "hey, we're Sun, nobody would ever
refuse us". -John
\_ Sun's proposal to replace uclink was pathetic; the people
who came to do the presentation had no idea what was in
it, and no idea of the Berkeley environment. -tom
\_ it's like the mama spider whose babies eat her once their born
\_ As long as someone eventually beats Micro$oft, I don't care.
\_ and linux is going to do that?? hahahaah
\_ Incidentally, why do people rave about Ed Zander? What exactly
did he accomplish at Sun?
\_ Four sysadmins administer a 400 server webfarm, along with
a database server, email server, all the corporate servers
and all the networking equipment. We can do this because
we run all Sun hardware and software. Tell me a place that
runs linux and does this. -Sun fan
\_ Any place that runs Linux on an AS400, dorkus. It's called IBM,
they are going Linux, it's well supported, and runs like a
charm.
\_ You know, administering 100 (likely similarly configured)
machines is not really _that_ hard for any competent sysadmin as
long as they run some form of *nix.
\_ Sun is out to destroy itself. Sun Service's new trick is to come
in, declare all the SA's overpaid, and replace them with Sun Service
H1-B's making $20/hr. -ax
\_ It is more like.. if they get destroyed, it would be because of
their own stupidity. They should have seen the Linux/x86 threat
coming a long time ago. Granted, they're ARE embrassing Linux and
x86 now and have plans to open source Solaris. So, it's possible
they're not a toast yet. Though, their sales people STILL don't
understand that the time when they could have made thousands and
thousands of profit per machine sold are over. The other day I was
looking at their education promotions page, and the best deal for
"low-end" server is a dual USIIIi rack box that costs $7000. Say
what? I can get a Dell twice as fast with storage 5 times the sun
for half the price. If that's the best they can offer, I feel really
really sorry for them. |
| 2004/11/8-9 [Computer/SW/OS/Solaris] UID:34760 Activity:low |
10/8 On solaris, what do the two flags [+-] between the username and the
#blocks used represent. Sample...
Block limits
User used soft hard timeleft
foo -- 2104304 5000000 10000000
bar +- 21538656 20000000 220000000 2.4 days
^^
\_ Could they mean you are over one quota but not the other???
Doesn't the man page answer this?
\_ The plus I believe does mean over soft quota. I am guessing
the second plus would mean time has expired. Not in manpage. |
| 2004/8/9 [Computer/SW/OS/Solaris] UID:32787 Activity:high |
8/9 old sun ray's are cheap on ebay (<$50). do you need special sw to use
them or will they work w/ a properly installed solaris installation?
\_ is it really worth $50?
\_ what's a cheap altenrative? btw, <$50 includes one i saw for
$10.50. w/ monitor = $75.
\_ You need sunray server software, free download http://sun.com |
| 2004/8/4-5 [Computer/SW/OS/Solaris] UID:32696 Activity:very high |
8/4 I have a Solaris UFS (USPARC, not x86) formatted drive.
Can I access this safely with a 2.4.xx Linux kernel through
USB (going to put UFS ide drive into an IDE-USB convertor.
Read somewhere that this might not be possible.
\_ ro sort of works, provided you turn off logging under
solaris.
\_ What does solaris logging have to do with accessing it
under Linux? If I had logging turned on under Solaris
then if I transfer the drive over to Linux it won't work?
Sorry, this is somewhat confusing to me. Can you elaborate?
I also only need RO access because I'm going to wipe the
drive and dump it after (it's only a 4 gig drive). As
I understand it, there's some problems with endianess
but I would think a proper UFS driver under linux would
take care of that.
\_ this is obviously because bill gates has hidden away the
inner workings of his OS and is evil!
\_Huh? -OP
\_ See how you are? |
| 2004/6/25-27 [Computer/SW/OS/Misc, Computer/SW/OS/Solaris] UID:31015 Activity:nil |
6/25 If you are running ISC Dhcpd 3.0.1rc{12,13} consider upgrading
to rc14:
http://www.us-cert.gov/cas/techalerts/TA04-174A.html |
| 2004/6/16 [Computer/SW/OS/Linux, Computer/SW/OS/FreeBSD, Computer/SW/OS/Solaris] UID:30826 Activity:very high |
6/16 I just got a new dual proc xeon server that will be replacing an
old E250 running Solaris. Gentoo is looking pretty good to me, but
I don't have too much experience w/ Linux/BSD. Instead of igniting
a flamewar, does anyone have a good URL that might line up the various
x86 *nixes so that I might figure out which is best for me? I've used
Debian in the past, but that was a while back.
\_ What are you going to use it for? It may not matter which one. It
is likely that whatever you or the eventually sysadmin or end users
are most comfortable with is the 'best' OS for your purposes.
\_ In the same vein, are there any specific disadvantages to ReiserFS
over ext3? |
| 2004/5/13-14 [Computer/SW/OS/Solaris] UID:30207 Activity:high |
5/13 Solaris Q: a friend needs to run a 32 bit app on a 64 bit solaris
machine. Is there a way to figure out which mode the kernel is in,
and/or to change the mode? Sorry for the vagueness, I'm just wondering
if there's a really easy solution. Thanks
\_ man isainfo. Note: you can run 32-bit apps on 64-bit solaris.
You can also easily google this shit. |
| 2004/5/11 [Computer/SW/OS/Solaris] UID:30146 Activity:very high |
5/10 I have years of experience with Solaris, but someone is asking if i've
used "Solaris availability enhancements". What the heck are they
\- these used to be known as Solaris Implants.
talking about? tnx,
\_ Means that the person you are talking to has never used Solaris.
There's no such thing as "Solaris availability enhancements" unless
person is talking about required patch clusters and server lockdown
procedures plus potentially fallover switching to "enhance" uptime
of said server. But I've never met anyone inside or outside sun
or on sunmanagers that has ever referred to anything of that sort.
-williamc
\_ Sun Clustering/HA stuff was very big not long ago. Haven't heard
much out of them of late because they've been pretty much priced
out by other HA options. The way it's phrased does indicate pointy-
hairedness, though.
\_ time to hit the google sonny
\_ Probably things like Sun Cluster and/or Veritas Cluster, Veritas or
Solaris LVM, etc.
\_ Note how most of the replies are in the form of "the idiot who
asked you probably is asking about X,Y,Z". There is no such thing.
This is some marketing person, pseudo-technical manager, or worse,
a marketing manager who once managed to turn on their own PC, right?
[restored, but maybe not completely intact] |
| 2004/5/3-4 [Recreation/Activities, Computer/SW/OS/Solaris] UID:29955 Activity:very high |
5/2 Is anybody running SpamAssassin on SOLARIS (5.7)?
Does it run reliably? Easy/Hard to install?
\_ I've run it on Solaris 5.8. Everything runs fine except for spamd
which segfaults on startup. I'm not sure why and haven't had the
time to look into it any further. Though this begs the question,
why are you still running 5.7? 5.8 is pretty crufty and ancient, I
can only imagine what 5.7 is like. -dans
\_ I've run spamd for long periods on Solaris 8 before, and it's
been pretty reliable.
\_ 5.8 isn't much different from 5.7. Many old 5.7 sites saw no
reason to up to 5.8 because, well, there aren't any. 5.9 added
a few things for some people. 5.10 might actually be new. For
the average joe running spamassassin on some old box he might as
well be running anything from 2.51 to 2.9 for all it matters.
\- anybody running crm114? having dealt with a reasonable
number of programs that can be described as "do complicated
analysis on hostile input", SA makes me nervous ...
especially when your are getting the arb input both in terms
of mail and users writing fairly arbitrary "policy". --psb
\_ I've been running crm114 for about 2 months but only doing
a small amount of training. It does at least as well as
spamassassin even though I'm not training it like I should
be. If SA catches it, crm114 mostly will. crm114 catches
many things SA doesn't. Unfortunately, a few things still
slip by *both*. Yes, I'm using both on the same stream. |
| 2004/4/20 [Computer/SW/OS/Solaris, Computer/Rants] UID:13299 Activity:nil |
4/20 I just learned that Sun Microsystems signed a non-aggression
pact with the Evil Empire Micro$oft. Since when did McNealy
becomes Bill Gates' bitch?
\_ "Keep your friends close, but your enemies closer."
\_ This happened a couple of weeks back. |
| 2004/4/15-16 [Computer/SW/OS/FreeBSD, Computer/SW/OS/Solaris, Computer/SW/Unix] UID:13219 Activity:high |
4/15 How do people deal with long file names when burning CD? Rockridge
extension only allows upto 32 (31?) characters.
\_ Rockridge supports up to 127 characters in UNIX.
\_ tar (or use whichever archiver) first
\_ can unix read PC and hybrid cd formats?
\_ Depends on the UNIX. Linux can red Joliet if you compile it in.
I'm sure *BSD has a backport. No go on Solaris.
\_ I usually burn w/ joliet extensions turned on. This results in
the shorter 8.3 file names on solaris boxes that can't read joliet,
but the file names show up correctly on Linux, MacOS (9/X) and
FreeBSD.
\_ I usually burn with joliet (-J), rockridge (-R/-r), the
translation table (-T), and long file names (-l). Probably
overkill, but no problems reading the names.
\_ cool, but which program on Darwin/OSX does that?
\_ hdiutil, toast and mkisofs can all do this (the
options specified look like they are for mkisofs)
To make a joliet/iso9660 hybrid using hdiutil:
$ hdiutil makehybrid -verbose -iso -joliet \
-default-volume-name [vol name] -o [iso file] \
[dir]
\_ mkisof is not available on Darwin. hdiutil only
does iso level 2 so probably cannot do the job.
\_ mkisof is not available on Darwin. Anyway hdiutil
to make an image before burning. Unfortunately if
you want Rock Ridge extension, you have to use
something else to make the image first. At least
for the version I have: 5.2
\_ http://csua.org/u/6xb?arstechnica.com
probably does the job though I haven't checked.
What's nice about Toast is that it does not need
to make an image before burning, unfortunately unless
you want Rock Ridge extension, and it cannot be operated
from command line. At least for the version I have: 5.2 |
| 2004/3/8 [Computer/SW/OS/Solaris] UID:12561 Activity:high |
3/8 Are there any commands on soda equivalent to prtinfo/psrinfo/isainfo
on SunOS 5? Thanks.
\_ The correct term is "Solaris", not "SunOS 5".
\_ whats sunos5?
\_ SunOS foo 5.7 Generic_106541-08 sun4u sparc
\_ That'd be Solaris 7.
\_ It would also be SunOS 5.7.
\_ sure, if you want to sound like a newbie.
\_ We went from SunOS 4 to Solaris 2.x to Solaris x
and you think 5.7 sounds like a newby? You're
the newby.
\_ http://wwws.sun.com/software/solaris/fcc/fcc.html
http://en.wikipedia.org/wiki/SunOS
okay, newbie?
\_ Using what other people in the field use makes you
!newbie. Sticking to some oddball strict number
scheme which has no more value than any other
oddball numbering scheme makes you stupid. I go
back far enough that I don't even want to tell you
what my first *nix was. I say Solaris 2.x --or--
Solaris x where x = 7,8,9,10,etc. Why? Because
people understand wtf I'm talking about.
\_ I understand. OTOH, the OS itself
identifies itself as SunOS 5.7. To say one
is a newby for calling the OS what it is
is the sort of thing only a newby would say.
I agree I also use Solaris more often. BFD.
\_ I'm not calling anyone a newbie but it
does sound 'funny' when I sometimes hear
someone say 5.7. Usually it is a recruiter.
I know a helluva lot about Solaris but I
still meet people who have been around even
longer than me or used it more extensively
or whatever who know things I don't. I do
not call anyone 'newbie'. Some people are
more experienced than me and some are less.
\_ dont be a dumb ass. |
| 2003/12/31 [Computer/SW/OS/Solaris] UID:11622 Activity:nil |
12/30 Does anyone know of a practical way to find out the sequence of
calls to functions from dynamically-linked libraries in a program
that doesn't have debug info? This is on solaris 8, btw. Also,
is there a utility that will show all the functions in a static or
shared library (on unix in general)? Thanks.
\_ "truss -u library program" will show you the function calls;
"nm library.so" will list the functions.
\_ thanks! i always assumed truss only did syscalls. |
| 2003/11/18 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:29640 Activity:very high |
11/17 What's the Unix command to see how fast the CPU is?
\_ dmesg
\_ I just tried this on Soda and it didn't work, just got:
dmesg: Command not found.
\_ on Linux: cat /proc/cpuinfo or x86info
\_ /var/run/dmesg.boot
\_ OK, so what's the real command to see how fast the CPU is? -op
\_ What exactly are you looking for thet the above do not provide?
\_ it would help if you mentioned which unix you were using. in case
you didn't know, it depends.
\_ Sun, Solaris -op |
| 2003/11/11-12 [Computer/SW/Languages/Perl, Computer/SW/OS/Solaris, Computer/SW/Unix] UID:11030 Activity:low |
11/11 ls -lt lists all files in a directory in chronological order.
For files nearly a year old, it only lists the date, but not the
time, of last modification. But it lists them in non-alphabetical
order. Thus, I suspect it is indeed listing them in order of most
recently modified. So, then, it presumably knows the modification
time. How can I get it to list the modification time, rather than
the year, for older files?
\_ get the ls that comes with GNU coreutils. it shows both date and
time, but only up to hour and minute.
\_ Try -T, works on FreeBSD
\_ Whoa, that's really useful, never knew about that; thanks! Too
bad it's not standard. -!op
\_ works here, but not on my machine at work.
\_ which runs....?
\_ Solaris
\_ Solaris follows the POSIX standard
\_ Maybe try /usr/xpg/bin/ls ?
\_ Try tarring up the file and scping it to soda
\_ As with all problems, this can be solved with perl: alias ls-lt to:
perl -e 'chdir$ARGV[0]if@ARGV;opendir D,".";print scalar localtime $_\
->[0]," $_->[1]\n"for sort{$a->[0]<=>$b->[0]}map{[(stat)[9],$_]}grep\!/^\../,readdir D'
(note the \'ed ! to make your shell happy). Then you can just say
% ls-lt /tmp
% ls-lt
etc.
and get the full dates. --dbushong
\_ or you might even be able to do tar tf on the tarfile |
| 2003/11/3-4 [Computer/SW/OS/Solaris] UID:10919 Activity:high |
11/3 I'm getting a no space left on device error even though all
my partitions have plenty of free space. Any ideas?
\_ you might have a large unlinked file that's still open for writing.
You can reboot to fix this or use lsof to find which process is
using that file.
\_ not sure what/how your OS reports it but you may want to check
your inodes.
\_ Solaris 9, my system hasn't been rebooted in 240 days. A Sun
guy told me switch to directio, but I don't see how that has
anything to do this with.
\_ Solaris 9, my system rebooted last week. thanks.
\_ Check the write permission on the devices.
\_ sun support just isn't what it used to be... directio?
You can reboot to fix this or use lsof to find which process is using
that file.
\_ Check the write permission on the devices.
\_ Check inode count like the above person said. |
| 2003/10/27-28 [Computer/SW/OS/Solaris] UID:10810 Activity:nil |
10/27 Where can I get a list of Solaris exit code meanings?
I am interested in 9, 10, and 11... Thanks!! I know my
program is not returning those, so something is setting it..
\_ /usr/include/errno.h
\_ exit codes, not error codes
\_ man siginfo and better yet: /usr/include/sys/signal.h
\_ sigh
\_ wow. both answers totally miss the mark. clever!
here's my attempt:
"\_ try looking at /etc/fstab". -ali.
\_ I can do better, "did you check /etc/nsswitch.conf?"
\_ /usr/pub/ascii
\_ good one, but to be truly great you'd have to direct
the op to the octal section.
\_ /usr/include/sys/errno.h
Why do you give an answer if you don't really know it?
\_ exit codes, not error codes
\_ man program, look under heading called EXIT STATUS |
| 2003/10/21-27 [Computer/SW/OS/Solaris] UID:10711 Activity:nil |
10/20 Looking for Sr. Unix sysadmin with strong Solaris, Veritas
NetBackup, Volume Manager/VCS skills. Experience with HDS and
Brocade a plus. Location just south of SF. E-mail me -- Marco |
| 2003/10/4-5 [Computer/SW/OS/Solaris] UID:10467 Activity:nil |
10/3 I'm a linux person, and my work is requiring me to learn Solaris,
which I have very little experience with. Any good books that
you guys can recommend or courses offered in Bay Area? I need
to learn Solaris 2.6 and 9, with emphasis on 9. Thanks.
\_ planning skills are easily transferable. hardware references can
be found online. I'd suggest picking up O'Reilly Essential System
Administration. Lots of translation tables between "the Linux Way",
"The Solaris Way", and so on.. BSD's, AIX, HP even. It's all the
same beast. Also, there's plenty of solaris fu on wall. throw out
a question sometime.
\_ They're all the same for the most part. Are you just being a low
level admin or they want you to run a veritas clustered database
on multiple E10Ks? |
| 2003/10/1-2 [Computer/SW/OS/Solaris] UID:10403 Activity:nil |
10/1 I'm on solaris and trying to increase the number of open file descriptors
my processes can have. I've used limit and ulimit under zsh and bash,
respectively, to set these to say, 400. limit and ulimit -a apparently
confirm these changes. But when I write a program to test this out,
it still fails after the default of 256 open files. Any idea what I
might be doing wrong? Thanks.
\_ The number 256 should've clued you in on what's going on.
\_ umm, duh. the question is why is it still at 256...
\_ yes, and you apparently missed the boat by a mile. 256 is
2^8. You're calling a clib function. Think about it. It's not
that hard....
\_ umm, again, no. you're probably missing the boat by
several miles then. do you even know c? or perhaps you're
just way over my head, in which case you should explain
to mere mortals what gems can be found in your seemingly
idiotic comments.
\_ Each time a file is opened space is allocated for
a new file descriptor. The descriptors are in an array.
In order to find an empty slot in the descriptor array the
kernel has to traverse the list. During the early days of SunOS
this was found to be rather expensive. So instead a bit field was devised
in order to determine used and open descriptor slots.
And guess what they used for the bit field, a byte.
Since process info has to be stored in kernel memory,
this was seen as saving space, so each process had
an associated bit field for the descriptor array.
Hence the 256 limit.
\_ welcome to a new century. the default hard limit is
now 1024. see below, limit is 256 only for standard
library functions, and this is removed on 64 bit
solaris 7. go figure. and i did say solaris, not
old SunOS.
\_ Solaris max is 256 for 32-bit apps. See rlim_fd_max in
http://docs.sun.com/db/doc/816-0607/6m735r5ev?a=view
\_ grr. well, that's not quite true. it turns out to be 256 for
standard library calls, like fopen. you can do 1024 using open().
unfortunately, python seems to use fopen. |
| 2003/8/28-29 [Computer/SW/OS/Linux, Computer/SW/OS/Solaris] UID:29505 Activity:high |
8/28 Followup on a previous post about network transfer of large files
and checksums: I have compared the 2nd download which passed the
md5sum with the first one which didn't. They have identical sizes
but differ in content on about 200 bytes out of about 640MB. Is
there a way to estimate the likelihood that this is the result of bad
transmission or a malicious substitution? I am asking both for
theoretical curiousity and practical interest. So besides some
high brow math. argument, is there some obvious indication like
whether the differences are concentrated, continuous, etc to check?
\_ Mount the iso file (assuming it's on a linux box) and poke around.
mount -o loop -t iso9660 filename.iso /mnt/tmp
\- yes there is a way to guess whether it is random or malicious
depending on what the contents are [probably], but it is a lot
of work, so i wouldnt bother. 200bytes is a hell of a lot.
that is a little strange. my guess is linux -> ass. --psb
\_ Have you determined what the differences are?
\_ All I did was was comparing the two images byte by byte with
a simple c program. Of course one could recursively look into
each volumes, and to be comprensive one has to look at
the partition map, catalog file, and auxillary partitions.
But as the posters above wrote, it is way TOO MUCH work for a
mild curiosity. I was asking if some statistical/probabilistic
analysis is possible (in theory) and some rule-of-thumb
available exists in practice. The transport was thru ftp, btw.
\- tcp checksum is not going to miss 200bytes in a <1gig xfer.
what you should do is do the xfer 100 times [or whatever]
and see how many times a strong checksum fails. if you do
that, i'd appreciate it if you would send me the info.
linux has a history of flailing on large data. --psb
\_ I transfer 8GB disk images and 600MB iso's between
my linux boxes. I've never had any problems. what do
you mean linux "flailing on large data"?
\_ I throw around 2 terabytes of data with linux every other day
and I haven't noticed any data loss yet but I have not
conducted an exhaustive statistical study. - danh
\- do you guys actually check the data or do you cross your
fingers? obviously if you dont look, you wont find.
also it may not manifest itself withing a certain range
of behavior/configurations.
anyway, first hand, i have had linux system writing
corrupted packets on the the net [went away when ethernet
driver was changed]. when we changed various things in
bpf and syskonnect ethernet driver fleebsd was fine
with our hacks, linux occasionally had issues (we didnt do
too much research on what the problem was ... we just abandoned
it ... and the problems seem to in part go away when we had
faster processors and faster disk bus). i dont remember which
file system it was, but one of them lost us some data and it
didnt appear to be a hardware problem [was a while ago also...
lately i havent been looking but havent casually noticed
data loss at fs level]. i dont need to say anything about
linux nfs server. admittedly these are rare, but they are
in areas you expect perfection. a bigger problem is just
general "weird behavior" under load [or sometimes even
not under load]. linux does too many short cut things for
"typical case" speed hacks. this can lead to your being out
to sea when something goes wrong [e.g. when you look at a
solaris crash dump, you have much better info than trying
to figure out what happened in the linux case. this might
partly be my better knowledge of solaris but in some cases
the relevant info about the thread state, locks, watchdogs
simply were not there] and also the system behavior often is
sort of unusual under load [e.g. low free memory + high io,
compared to FreeBSD and solaris (although when various large
changes were made in solaris kernel algorithms for short
periods i did see some performace issues)]. finally i dont like
the way the memory-file system subsystem has been evolving.
recently seen some problems in work environments with lots of
(tcp) connections ... you get weird hangs on clients when the
server drops packets ... admittedly this might have been fixable
by throwing hardware at the problem or tweaking various para-
meters (and this was on some HPC enviornments were we could not
compare against solaris/bsd).
YMWTGF: andrew hume HotOS linux suspect --psb
\_ Our answer was much simpler than yours. After too many
lost files, NFS problems, dropped packets, etc, etc, we
simply stopped using Linux because it sucks. We didn't
have the time to get into this driver vs that driver or
what kernel patch might have helped or which NIC, etc.
Linux = not ready for enterprise = out the fucking window.
Staff time is more expensive than the value of possibly
finding a solution to kludge Linux into working. The
moment we switched to real OS's our problems just magically
went away without hiring a team of Linux kernel developers.
Linux is cute but their development philosophy precludes
it's use in enterprise environments. Just FYI, I'm tossing
around 20-30TB/month between various hosts.
\_ Which OS did you switch to? FreeBSD? Solaris?
\_ Yeah, especially now that Sun sells the X1 for under $1k.
\_ I should have added: The system from which I run the ftp was
OS X, which is a (free)bsd derivative. And I also noticed that
the bad download had wrong modification time. It was set to be
the day of the download, even though I have "preserve" on. |
| 2003/8/1-4 [Computer/SW/OS/Solaris] UID:29206 Activity:moderate |
8/1 Snapfish engineering was really happy with the Berkeley resumes
that came in for the junior java position. We're now hiring a
mid level unix admin. It's a 24x7 job in SF working in my group.
Roughly 100 Solaris/Linux boxes and 90 terabytes storage in production.
Cal degree a plus. Inexperience *not* a plus this time, sorry. Email
me with questions and resumes. Yes, it's a pager job. That's what
24x7 means. -reiffin
http://www.craigslist.org/sfc/sad/14320326.html
\_ no day off for you!
Roughly 100 Solaris/Linux boxes and 90 terabytes storage in production.
Cal degree a plus. Inexperience *not* a plus this time, sorry. Email
me with questions and resumes. -reiffin
http://www.craigslist.org/sfc/sad/14320326.html
\_ Run from the pager.
\_ so is it a trend now to hire more junior positions (which are
plentiful & cheap) than a few skilled positions due to the fact
that within the past 30 years computer science has been studying
how complex tasks could be abstracted, decomposed, and
repartitioned into smaller, simple, and more manageable tasks?
\_ he just said midlevel, not junior. go away.
\_ Not to mention finding idiot juniors who are perfectly happy
to be saddled with a 24x7 pager with a 15 minute turnaround.
I seriously hope you don't have just one person on that sort
of pager schedule.
\_ Really, a single person on 24x7 pager? Sounds like a job
description written by someone who's never carried one. You'll
need at least two people, or the poor bastard will burn out in
< 1 month.
\_ It isn't the only person on the pager. It's a rotation. -reiffin
\_ You troll so stupid, you make reiffin look smart. ;-) |
| 2003/7/4-5 [Politics/Foreign/Europe, Computer/SW/OS/Solaris] UID:28923 Activity:nil |
7/3 The BLOB: http://news.bbc.co.uk/2/hi/science/nature/3039102.stm New Solaris System found: http://news.bbc.co.uk/1/hi/sci/tech/3041220.stm |
| 2003/6/29-30 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:28859 Activity:kinda low |
6/28 is there a unix instant messenger client for aol instant messenger
stuff that supports file transfer?
\_ I believe gaim now has support.
\_ I used to use a tcl/tk version of Aim client on Solaris.
that was years ago. GAIM has the most active development.
Have you also tried Everybuddy? --kngharv |
| 2003/6/24-26 [Computer/SW/OS/OsX, Computer/SW/OS/Solaris] UID:28828 Activity:high |
6/24 About Apple's 64-bit claim. Does Apple ship a 64-bit OS or does it
support running 64-bit applications right now?
\_ when the G5 ships with 'Smeagol' 10.2.7, it will support
G5-optimized apps and presumably 64-bit apps.
\_ If you look at the commits for darwin, it doesn't
look like the kernel/libc/userland have been updated
for 64-bit support. A select few apps might get 64
bit support, most apps will still be 32bit for the
next year or more.
\_ god, i hate apple. i wish Tolkein were alive to sue them,
and hated them as much as i do.
\_ For a code-name that isn't supposed to be public?
Don't hate the player, hate the game!
\_ Why not Carl Sagan (aka BHA) did it.
\_ can anyone hate them as much as you do, my precious?
\_ They're based in Cupertino, CA. Good enough reason for
me to like 'em, they pay CA taxes. Maybe you'd rather
live in Texas? Don't let the door hit you on the way out.
\_ The 970 is a 64-bit proc. OS X 10.2.x isn't 64-bit clean.
It looks like 10.2.7 and 10.3.x will be able to run 64-bit
apps but the kernel/libc/userland will probably not be 64-bit
clean till 10.4.x.
\_ In short: no. It's all promises with unknown future performance.
\_ Of course, it probably wasn't unexpected. Take a look at Sun's
example. The UltraSPARC I (Sun's first 64-bit CPU) came out
first in 1995 or so while Solaris 7, the first 64-bit Solaris
release, came out only in 1998 or so. |
| 2003/5/14-15 [Computer/SW/OS/Solaris] UID:28440 Activity:nil |
5/14 Where can I get a color xterm for Solaris (sparc)?
\_ Upgrade to Solaris 9 or compile the xterm from XFree86 distribution.
\_ Has anyone had success compiling the XFree86 distribution on
solaris? I can't get very far on it.
\_ works fine for me, but I've had to learn far more about
imake than you ever want to know. -alan-
\_ Also available for Solaris 8 on the FreeWare Companion CD from
Sun. Probably also on http://blastwave.org & <DEAD>sunfreeware.com<DEAD> |
| 2003/1/16-17 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:27121 Activity:very high |
1.15 cat t | grep -v FOOM > t AND
works fine on BSD but not on Solaris. Is this a
a) race condition, b) platform dependent issue, c) Solaris bug ?
so what's the easiest way to do that on solaris?
btw, grep -v FOOM < t > t doesn't work on either.
\_ b)
\_ Isn't that a shell, rather than platform, issue?
\_ Depending on your shell, the shell will call open with
O_WRONLY|O_CREAT|O_TRUNC when you specify >. And > will
get eval'ed before |.
\_ Depending on your shell, the shell may call open with
O_WRONLY|O_CREAT|O_TRUNC when you specify > and it
may eval > t before eval'ing | grep.
\_ You're all wrong. It's user error! Duh! |
| 2002/12/17-18 [Computer/SW/OS/Solaris] UID:26837 Activity:moderate |
12/17 Does anyone know where I can find source code and/or executable
for sar for Solaris 2.9? I couldn't find it on google. Thanks!
\_ Do you have a support contract from Sun?
\_ Executable: SUNWaccu on Solaris 9 install media.
Source: If *.edu, send $100 to Sun. If not, send many times
that to Sun. -alan- |
| 2002/12/5-6 [Computer/SW/OS/Solaris] UID:26725 Activity:high |
12/5 Any idea if there's a limit to the number of mount points an nfs
client is allowed to have in either linux (any) or solaris (any)?
I've got data all over the place and need each client to mount a *lot*
of servers. Thanks!
\- not practically.
\_ In the old days the max number of mount points was around 256.
These days its probably a lot larger. If you are worried about
this, the best solution would be to use automount. This would
\- i have reasonably large
automounter maps ... i see
automounter on solaris
grow over time to 100meg+
in size, so i could believe
there is some problem with
it that may manifest itself
with a huge number of
mounts. --psb
\_ Yeah, I've seen this
too, but if OP wants
to limit the number
of mount points used
on ave. then automounter
is the only real soln.
The mounts are being hit all the time by various apps so _/
automount won't really help. I just need to know if there's
a real limit and how far away I might be. I've got 50+ now
and was going to be maybe be adding another 50+ over the next
2-3 months and then more after that every so often forever. I
need to know in advance if I should be telling management they
can't do what they're wanting to do or not. Thanks! --OP
\_ You may be able to add lots of mount points but at some
stage beyond 200 or so you will start to see stability
and performance issues (at least with older versions of
linux, 2.0 and 2.2 kernels, and Solaris, 2.5, 2.6 and maybe
even 7). In the long run your best bet will be to ditch
nfs and to use something like afs which was designed to
handle your type of problem.
limit the ave. number of mount points in use at any given time.
\_ Thanks! |
| 2002/11/15 [Computer/SW/OS/Solaris] UID:26555 Activity:nil |
11/15 Does solaris have a iso-8859-1 to UTF-8 character converter?
I'd like something that takes an accented "a" and converts it to
á Thanks. |
| 2002/11/2-3 [Computer/SW/OS/Solaris] UID:26382 Activity:kinda low |
11/1 Solaris packages. Anyway to install multiple instances of the exact
same package? Same arch, version, etc. Just different install dirs.
\- you could use something like the modules package. last i heard
\_ the modules package?? -op
it was being worked on by a fellow at sun but i think it may have
been abandoned/rewritten. you can look for the usenix paper. --psb
\- of if youa re talking about packages as in "/opt" the
module stuff doenst really apply. i thought you meant what
is a convenient way of having an emacs 19/20/21 or mutiple
gcc instances etc. --psb
\_ I suggest tweaking the package by hand. What package are you trying
to dual install and why? Use softlinks.
\_ pkgadd -R
\_ fucker, stop posting man page answers. dick. what fun is that?
\_ Should I just say RTFM insted? I thought about mocking psb
for coming up with a stupid and complicated answer to a
simple question, but I thought the answer was funny enough.
\_ Ok, ok, you get a pass on this one. |
| 2002/10/25-28 [Computer/SW/OS/FreeBSD, Computer/SW/OS/Solaris] UID:26322 Activity:low |
10/24 Anyone know of a way on solaris to see how many bytes went out
on a particular interface? Analogous to netstat -ib of freebsd?
I have some tools to do this, but I am wondering if there is a
stock way. --psb
\_ check out orca or setools. Let me know if this works for you.
Or ucd-snmp or netsnmp. -abe
\_ that's not what he asked.
\_ I'll kill you and eat your children. -- Cthulhu
\_ You're sleeping. Anyway I don't have children and I ran
out of SAN years ago. |
| 2002/10/18 [Computer/SW/WWW/Browsers, Computer/SW/OS/Solaris] UID:26248 Activity:nil |
10/18 How do I tell plain old Netscape 4.76 for Solaris to not allow
any pop-up windows?
\_ I think you upgrade to a newer version of Netscape. Or you
could try the Solaris version of Opera, the fastest browser
on earth.
http://www.opera.com/products/desktop/index.dml?platform=solaris
\_ Disable java entirely or install mozilla1.1.
\_ javascript, twink |
| 2002/9/12 [Computer/SW/OS/Solaris] UID:25855 Activity:high |
9/11 On SunOS 5, how do I use the -k option in "sort" to specify multiple
sort fields? I tried "-k start1,stop1 -kstart2,stop2",
"-k start1,stop1+start2,stop2" and "-k start1,stop1 start2,stop2" but
none works. The man page doesn't descript this. Thanks.
\_ I never got -k working sanely. I use the +n format just fine. I
use perl if I get desperate. |
| 2002/8/26 [Computer/SW/OS/Solaris] UID:25688 Activity:high |
8/26 How do you specify the defaultrouter on the second interface, say
qe1, on Solaris?
\_ ifconfig manually or via script |
| 2002/7/14-15 [Computer/SW/OS/Solaris] UID:25353 Activity:moderate |
7/13 Has anyone experienced the difference in qsort between Solaris and
Linux? I found that when the indices are the same, the sorting order
is different. Is there any way to get the Solaris-equivalent qsort
on Linux?
\_ The recursive nature of qsort means that you can never get
a stable sort out of it (AFAIK), meaning that
"The relative order in the output of two items that compare
as equal is unpredictable." (-Solaris manpage). Use mergesort.
\_ The strange thing is, the qsort on Solaris and HPUX give the same
sorting order. Is there any way to get the source code of qsort
for HP or Solaris? -op |
| 2002/6/27-28 [Computer/SW/OS/Solaris] UID:25230 Activity:nil 66%like:25224 |
6/27 Where to get the "screen" source code or binary for Solaris?
\_ http://www.sunfreeware.com
\_ great, now we have "where to" questions in addition to the "how to"
ones...
\_ http://freshmeat.net/search/?q=screen§ion=projects |
| 2002/6/27 [Computer/SW/OS/Solaris] UID:25224 Activity:nil 53%like:25223 66%like:25230 |
6/27 Where to get the "screen" source code or binary for Solaris?
(-=-= One MOTD, Under PSB, Indivisible, with Liberty and Justice for None =-=-)
\_ Well there's sure as hell no justice around here.... ;-)
And shouldn't that be "Under G-d" according to the recent motd
search/replace job someone did on the motd? |
| 2002/5/28 [Computer/SW/OS/Solaris] UID:24962 Activity:high |
5/28 Why would I "buy Solaris" instead of getting the free version?
http://wwws.sun.com/software/solaris/get.html
Do I need to pay the extra $200 to keep my support?
It looks like I need to "buy" Solaris only if I have multiple CPU's.
Am I missing anything?
\_ Say what?
\_ It looks like you have to purchase a media kit and licenses
separately. There are several media kits that differ by the amount
of included printed documentation, value-added software, languages
CDs, etc. I just ordered a Solaris 8 02/02 media kit with all
software but no printed documentation for $75 and Solaris 9 FCS
media kit for $95. Now, if you need to run it on multiprocessor
machines, add the cost of license on top of that. As for the "free"
version, it only includes the Solaris OS and no value added
software. May be this is exactly what you need. Solaris 8 media
kit includes some useful freebies like the iPLanet directory server
with a license for up to 200,000 directory entries. Note that
Solaris 9 ISOs are still not up for a free download..
\_ I think that you also get the vol mgr with the paid version
\_ That was up to Solaris 7. Disksuite packages for Solaris 8 were
included on the second Solaris software CD, although, hidden in
a directory suspiciously called "EA". As of Solaris 9, the vol
manager has been integrated into the OS, so you shouldn't even
need to deal separate packages any more. |
| 2002/5/21-22 [Computer/SW/OS/FreeBSD, Computer/SW/OS/Solaris] UID:24899 Activity:insanely high |
5/20 Anyone have any luck with Sparc64 + FreeBSD?
Do you recommend it?
\_ God No! Are you insane???
\_ OpenBSD Sparc64 is mostly usable. FreeBSD Sparc64
barely runs
\_ Why not Solaris?
\_ commercial software is wrong. --cowboy
\_ wrong? take it to the church, son.
\_ But was it included when you bought the Sparc?
\_ Solaris 8 is freely downloadable or $75 for a media kit.
\_ "yeah but i cant get the source and make my own solaris
distro so it doesnt count! how will i fix bugs and
release them to the community? i mean like i know i
dont even code and couldnt find a bug if it bit my
nose off but yeah!"
\_ And releasing source code will endanger our
soldiers in Afganistan! --msft
\_ most of Solaris 8/FCS source code is available for
download. You are free to fix the bugs but you must
submit them to Sun. If your organization asks nicely
you could get the source for other releases too.
\_ "most? thats evil! i want the whole source and
the right to make my own kewl-d00de d1st0! awl
0p3n s0urc3 un1t3!!!! d00dez!!1"
\_ sun wants to make everything available, but
the patent/copyright holders on parts of
the kernel and userland won't let them.
\_ "if sun cared about open source they'd
rec0de all the evil parts and release
as GPL! Death to Sun!"
\_ been running debian sparc on an ultra 1. works fine. (potato)
\_ How about Open vs Net ?
\_ Real sparc64 (HAL/Fujitsu) or UltraSparc?
\_ hahahahahh.... HaL... Real.... sparc64....
ow...
\_ Well, if it's SPARC64[tm], it must be HaL. Unfortunately
some stupid opensource people use "sparc64" when they really
mean UltraSPARC[tm].
\_ Whee.. I worked at HaL, back when they were trying to
rush their 2.5.1 out the door while Sun was prepping
Solaris 7. Neat boxes, though. --scotsman
\_ I have an E250? Is that "real" or just ultrasparc? Where
is the confusion coming from? Who invented the "fake"
sparc64?
\_ HaL was to be Fujitsu's big bid in the processor market.
They grabbed the trademark "sparc64" out from under Sun,
seriously pissing them off. The hardware was really
pretty cool. At the time, one of the few things around
that could take 4G of RAM. They just couldn't get the
OS dev schedule up fast enough to compete against Sun
at their own game. --scotsman
\_ Doesn't sparc64 run a stock Sun Microsystems version
of Solaris?
\_ not at the time. (this was in like '97. I dunno
what they run now, or even if sparc64 machines
are still being built) --scotsman
\_ Strangely enough, http://daily.daemonnews.org just reports that the
Sparc64 port is "fully self-hosting"-- meaning it can actually
compile itself and run. Short answer-- it's not ready yet. And
it doesn't work with YOUR cable modem. |
| 2002/5/7-8 [Computer/SW/Languages/Misc, Computer/SW/OS/Solaris] UID:24745 Activity:high 54%like:23760 |
5/7 On XP I can do route print to see the routing table. How do I do that
on Solaris and Linux?
\_ netstat -r
\_ netstat -rn (if you don't want name resoultion)
Destination Gateway Genmask Flags MSS Window irtt Iface
10.2.0.3 0.0.0.0 255.255.255.255 UH 40 0 0 ppp0
10.2.0.3 0.0.0.0 255.255.255.255 UH 40 0 0 ppp1
\_ what does it mean when your gateway is 0.0.0.0? I don't get it
\_ It points to the other end of your ppp connection
\_ It means someone has hacked your computer |
| 2002/5/7-8 [Computer/SW/OS/Solaris] UID:24739 Activity:high |
5/7 What does /etc/tm do, on sunos? How do I search for this on google,
it keeps insisting "/etc/tm" is "etc tm"
\_ google 'solaris "etc tm"'
\_ On my solaris boxes it's an empty directory. I don't have a sunos
box. It isn't in the solaris man pages I have.
\_ rm, mv, or chmod it and see what breaks. |
| 2002/4/4-5 [Computer/SW/WWW/Server, Computer/SW/OS/Solaris] UID:24323 Activity:very high |
4/4 Is there a way to make Apache case-insenstive (on Solaris)?
(i.e., http://foo.com/cgi-bin/bar?param1=foobar should be
the same as http://foo.com/Cgi-bIn/BaR?param1=foobar; obvisouly,
I can write my cgi-bin's so that all the params are
case-insenstive, but the leading URIs?)
I have used google and have been on Apache's web site. Thx.
[...]
\_ Look, dummy, the answer is you can't do it. Your only
other choice is hacking the url parse code in apache
to lower case the entire URL. Good luck with your
coding project.
\_ Ok, thx. That's what I thought and I just
needed someone to confirm it.
\_ it's wrong. mod_speling does exactly what
you want. Try it, nimrod.
\_ why is "mod_speling" spelled with only one
L? Is it supposed to be some dumb attempt at
being humorous?
\_ yes. laugh a little!
\_ Wow this was tough to find. Took me about 15 seconds.
http://httpd.apache.org/docs/misc/FAQ-H.html#rewrite-nocase
You're welcome.
\_ errr...I have read that and the speling module. mod_speling
only makes the document name referenced case-insenstive, not
all the elements that construct the URI.
all the elements that construct the URI. From Apache:
"the module is unable to correct misspelled user names (as
in <DEAD>my.host/~apahce<DEAD> just file names or directory
names."
\_ Grasshopper, the wind blows through the trees yet disturbs
not the trunk, only the leaves....
\_ huh?
\_ Grasshopper, the answer lies before your eyes are
darkened by your own thoughts.
\_ it corrects directory names, can't you read?
\_ Run apache on windows.
\_ Ew. |
| 2002/3/22 [Computer/SW/OS/Solaris] UID:24187 Activity:nil |
3/21 Updates to ~scotsman/bin/patch_{dl,in}. If you have to deal with
Solaris patches, these may help make life less obnoxious. |
| 2002/2/25-26 [Computer/SW/Security, Computer/SW/OS/Solaris] UID:23970 Activity:high |
2/25 Is there anyone out who understands the NT security API? All I'm
trying to do is set permissions on a directory: Everyone group,
full control, inheritable by child objects and containers. Then I
need to know how to create files so that they don't override the
parent ACL. Should they have a NULL SD, or a default SD with a NULL
DACL?
What I'm doing now is setting security on every file create and
copy, which is error prone. CopyFile doesn't copy the SD, so I do
SetNamedSecurityInfo(DACL_SECURITY_INFORMATION|
PROTECTED_DACL_SECURITY_INFORMATION) on the new file. It would be
better if it just obeyed the parent directory settings. -sky
\_ I know how to do the similar on Solaris on but not NT. Sorry.
In Solaris I set the parent directories ACL and mask and then all
children (both files & directories) inherit ACL. At least when
you do commands like cp/cat/vi. Okay, so it's not going thru the
API so it's not similar thing. There are some oddities when on
older versions of Veritas products though. Are you using Veritas?
\_ Apparently no one understands the NT security API. What I _do_
know is it has nothing to do with the way Veritas or Solaris work. |
| 2002/2/12-13 [Computer/SW/OS/Solaris] UID:23853 Activity:moderate |
2/12 It's not in dmesg, prtdiat or prtconf or the eeprom variables: Is there
a command on solaris that may list the serial number? ISW?
\_ No. |
| 2002/2/8-9 [Computer/SW/OS/Solaris] UID:23815 Activity:high |
2/7 Is there a Solaris LKM or some other way to log all the filenames
which are read and written on a SUN ?
\_ how much will you pay me to write you one --aaron
\_ FIE DOLLA! The amount you paid for h07 a$|n ho in China
\_ the Dell dude is better looking than aaron [although the dude is
definitely more annoying]
\_ lsof. "ls open files". Open source. Get from net if not alreadyy
installed.
\_ That can only take snapshots. I think the poster means
what he means.
\_ On occasion, I use prex, which loses the filename
abstraction and is a bit of a pain to use, but it's
a logging system not a snapshot system. OK, thanks. -psb
\_ If there is no such module, it shouldn't be very hard to write as
long as you can find documentation on how to do it. Doing something
like this on Linux seems to be trivial after I have read a
related article on Linux Journal recently.
\_ It's better than what he has now and the source is open.
\_ um... he didn't say what he has now, and it doesn't do
what he wants. how is that "better"?
\_ If he had anything now he wouldn't be asking on the
motd and it kind of does what he wants. Since he
didn't say what he's using it for you don't know it
doesn't do what he wants. Out in the real world away
from the small minded ivory towers of academia we
call "something" better than "nothing". Until the
poster comes back and say exactly what they need "lsof"
is a better answer than "you're wrong because that's
not what they wanted". ok tnx.
\_ Most people who know what a "solaris lkm" is also
know what lsof is. The poster does mean what he
means. Thanks for the suggestion but dont get
snippy. This is not a academia vs industry thing.
Both communities need to take file system traces
so I am looking for such a tool.
\_ It isn't academia vs industry. It's "here's a
partial answer, go look at the source" vs. "you
didn't 100% answer the question so you're wrong".
We still don't seem to have the OP here to say
exactly what they know or don't know and want/need
or don't want/need.
\_ it can teach him the routines used.
\- on occasion i use prex, which loses the filename
abstraction and is a bit of a pain to use but it is
a logging system not a snapshot system. ok tnx. --psb
\_ ok tnx.
\- pls pay me the "ok tnx" tax. ok tnx. --psb
\_ The information wants to be free! Fascist! ok tnx. |
| 11/23 |