Migrations
Aug. 16th, 2005 09:10 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
sobriqueBugger.
There goes another one.
Started migrating an oracle server.
Found some patch prereqs were needed.
Installed 'em.
Then had a server that kernel panicced on boot up.
That was fun.
Thankfully it _is_ possible to backout kernel patching on sun. It's somewhat un-fun, when you're trying to do so on a set of volume managed disks since booting from CD, you don't have the meta databases available, so you're trying to bodge a disk set together, into a state where you can remove a dodgy patch, and then start the server.
So, 21:00 hits, and I'm back to square one again. Well, nearly. I've got some files to finish copying before I'm right there.
*clunk*.
Hear that? That's my 'giving up and going home threshold'.
I need those patches for it to work.
I _probably_ need to Recommended and Security patch cluster this machine. That'll take an hour or so. _Then_ I'll still have to copy 130Gb of data to a new disk, assuming that _does_ do the job with these patches I need.
I really can't find the enthusiasm to keep going at a task that I _know_ if it all goes well will take me past midnight.
Better yet, I've got to SAN attach _yet another_ kludgy old server with a wonky OS build tomorrow, as a priority because it's part of the backup testing on some other project.
I hate my job at the moment.
This, I'm sure, will pass. Cos usually it's kinda fun.
Lots of evenings and weekends, and 3 or 4 critical jobs on the go at ones however, really do stretch my sense of professionalism to it's limit.
Bugger this. I'm off for a beer.
There goes another one.
Started migrating an oracle server.
Found some patch prereqs were needed.
Installed 'em.
Then had a server that kernel panicced on boot up.
That was fun.
Thankfully it _is_ possible to backout kernel patching on sun. It's somewhat un-fun, when you're trying to do so on a set of volume managed disks since booting from CD, you don't have the meta databases available, so you're trying to bodge a disk set together, into a state where you can remove a dodgy patch, and then start the server.
So, 21:00 hits, and I'm back to square one again. Well, nearly. I've got some files to finish copying before I'm right there.
*clunk*.
Hear that? That's my 'giving up and going home threshold'.
I need those patches for it to work.
I _probably_ need to Recommended and Security patch cluster this machine. That'll take an hour or so. _Then_ I'll still have to copy 130Gb of data to a new disk, assuming that _does_ do the job with these patches I need.
I really can't find the enthusiasm to keep going at a task that I _know_ if it all goes well will take me past midnight.
Better yet, I've got to SAN attach _yet another_ kludgy old server with a wonky OS build tomorrow, as a priority because it's part of the backup testing on some other project.
I hate my job at the moment.
This, I'm sure, will pass. Cos usually it's kinda fun.
Lots of evenings and weekends, and 3 or 4 critical jobs on the go at ones however, really do stretch my sense of professionalism to it's limit.
Bugger this. I'm off for a beer.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2005-08-16 09:54 pm (UTC)http://www.redbrick.dcu.ie/~valen/epic.txt
no subject
Date: 2005-08-17 08:01 am (UTC)Solaris on Sun, I am still a big fan of.
I mean seriously, when you screw up the kernel on an NT machine, your _only_ option really is to rebuild.
In this case, rebuilding wasn't an option, so I was thankful for alternatives.