Gordon Messmer wrote:
>On Wed, 2002-10-09 at 18:21, Scott Bower wrote:
>
>
>>My biggest rush to
>>deployment was as a result of a major drive faltering and the fact that
>>RH8.0 contained JFS which would allow dirty great archives...
>>
>>
>
>What's a dirty great archive?
>
>
>
>>not working and other things segfaulting preventing anything at all
>>
>>
>...
>
>
>>fingered salute would work). On booting, the system replayed fs journals
>>and found them good however on automounting, something shit itself (read
>>Core Dumped) and the entire mount process stopped.
>>
>>
>...
>
>
>>noauto in fstab to allow the bloody thing to work and found that these
>>mounts worked fine when manually mounted (WTF!?!!).
>>
>>
>...
>
>
>>like random disappearing mozilla and things like that.
>>
>>
>
>Given the wide range and sporadic nature of the problems you've had, I'd
>say you were awful quick to stop considering hardware trouble.
>
>Red Hat Linux 8.0 has, in my opinion, been the best .0 release I've seen
>yet (used since 4.2, so I've seen 5.0, 6.0 and 7.0). I've not had any
>stability trouble with the final release; though I did during the last
>beta, so I'm watching :)
>
>
>
>
>
>
Well, I had no problems with this system for six months prior, the
kernel began reporting Timeouts on the device about the same time the
drive started belting the armature against the frame end... out it came
after installing RH8 and JFS and creating that 33Gig compressed tar
archive I was talking about, the drive was removed and replaced with
another unit which hasn't had any problems (logged or otherwise) however
the system has had plenty of it's own. I have been rectifying each
fault one at a time finding another as I go, after getting this system
up and manually mounting the JFS trees, all has been good (nervously
stroking wood...). I haven't had the gall to touch the bloody thing since.
the JFS Shares I created are on separate SCSI drives (failure unit was
an IDE item), the Mozilla problem has already been bugzilla'd by someone
else, the JFS/mount Segfault only happened when booting (perfectly every
time) and I seemed to have worked around it rather easily and have since
found it bulletproof in comparison to booting with automounting.
generally a hardware fault is sporadic not reproducable perfectly each
time, this generally (I have to admit) indicates a bug as opposed to a
hardware fault.
The root and usr directory incidentally were not on the disaster disk, I
know better than to blame the os for a bad system drive. The harware
has been interrogated at length over this issue and none of the problems
I've come accross have any hardware malice in them. Hardware faults
don't rename files to ~.rpmsave or leave messages in the log like "Read
this doco on upgrading the config, starting in useless mode" while
indicating at boot with [ OK ] and showing pids, many of the services
don't get verified until hours later when you require their use remotely...
I am also working with three SOE development systems at work and I have
had my fair share of problems there too, many of the faults I've come
accross have already wound up on Bugzilla and are yet to be fixed, none
of the systems have problems in their parallel OS installations and the
kernel certainly isn't reporting curious things in it's logs either. Not
to step on the arrogance button too much but I've been working with IT
for around twelve years, I'm currently working with the Australian
Dept. of Defence (that's how we spell it) as a WAN Engineer, I've done
my share of field work on PC Support and site management issues before
arriving here so please believe me when I say I've checked any and all
reasons for kit failures,hardware and otherwise. much of the failures of
my webservice at home are attributed to my selective ignorance of
correct deployment protocol , I had an opportunity to kill two birds
with one stone and silly me, I took it. The SOE Stuff is a different
story however and the problems we've had there are indicative that I
wasn't completely at fault.
--
redhat-list mailing list
unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe
https://listman.redhat.com/mailman/listinfo/redhat-list
