🚀 go-pugleaf

In <mazulauf-0903961351200001@grumpy.met.utah.edu> mazulauf@atmos.met.utah.edu (Mike Zulauf) writes:

>In article <danpop.826236568@rscernix>, danpop@mail.cern.ch (Dan Pop) wrote:
>
>> And my main point, which _you_ missed, was that a correctly administrated
>> Sun doesn't crash.  I backed my point with hard data, from workgroup
>> servers which support several users working from X terminals every day.
>> Their uptime is now in the forty-something range.
>
>Let me reiterate this.  I find SunOS/Solaris to be reasonably stable. 
>Certainly much more so than the MacOS.  But it can and does crash
>occasionally, even on 'correctly administrated' machines.  To deny this is
>foolishness.

Time to reiterate our small test:

cnwgs2:~:ROOT 24> rsh -l iss cms1 uptime
 11:44pm  up 44 day(s),  3:36,  1 user,  load average: 0.06, 0.09, 0.11
cnwgs2:~:ROOT 25> rsh -l iss cms2 uptime
 11:44pm  up 44 day(s),  6:20,  1 user,  load average: 0.09, 0.11, 0.13
cnwgs2:~:ROOT 26> rsh -l iss cms3 uptime
 11:44pm  up 45 day(s),  9:27,  3 users,  load average: 0.14, 0.09, 0.12
cnwgs2:~:ROOT 27> rsh -l iss cms4 uptime
 11:44pm  up 45 day(s),  9:24,  3 users,  load average: 0.05, 0.09, 0.14
cnwgs2:~:ROOT 28> rsh -l iss cms5 uptime
 11:44pm  up 44 day(s),  5:05,  4 users,  load average: 1.68, 1.76, 1.82
cnwgs2:~:ROOT 29> rsh -l iss cms6 uptime
 11:45pm  up 45 day(s),  5:56,  2 users,  load average: 0.05, 0.09, 0.11

Being midnight, the machines are idle (with one exception) but during
normal working hours, they do a lot of number crunching, apart from
email, netscape, etc.

If you'd give me a recipe to crash them, without root privileges, maybe
you'd have a point.

I'm not stupid enough to believe that the current version of Solaris
is completely bug-free, but these are heavily used machines, yet they've
been up since the last reboot, which was done 6.5 weeks ago for 
administrative reasons.  I didn't claim that they'd _never_ crash in any
circumstance (only a fool could make such a claim).  On the other hand,
I proved that these machines don't crash under their current usage as
workgroup servers.

Dan
--
Dan Pop
CERN, CN Division
Email: danpop@mail.cern.ch 
Mail:  CERN - PPE, Bat. 31 R-004, CH-1211 Geneve 23, Switzerland