Well, Sunday I was messing around with the systems at home. Installing updates, kicking the NFS server for fun, only to find that when I logged in to my workstation, KDE wouldn't start. I got a solid blue background and a cursor that I could move around, but nothing was happening. Nothing suspicious in top, ps, or even .xsession-errors per se. I just could not figure it out for the life of me.
So I started thinking, well, maybe it is just me. Some config file for KDE got corrupted and it is holding the whole thing up. So I started poking around, but couldn't see anything obvious. So I trashed some configs I figured might be the problem, still no change so I restored them. Then I figured I should test whether it was a config thing so I stomped the whole config directory in .kde, still not change. I was still convinced at that point that it was just me and in particular, something with KDE, so I stomped my whole .kde folder. Still no change.
I had to admit I had assumed I knew what was wrong when clearly, I was just making an ass of u and me... Mostly just me.
So then I tested whether other user accounts were affected. Sure enough, other accounts were affected. Precisely, the network accounts, but not the local accounts. Something was wrong with NFS.
I poked at siona a bunch and rebooted the workstation a bunch, but still no change. Every time, friday would reboot and I could authenticate, but then nothing would happen on login. Well, not quite nothing. It was just so slow that login/logout took something like a half hour.
Finally admitting I could not fix the problem by hand, I installed the latest kernel on siona (the only good reason for rebooting a GNU/Linux box other then adding new hardware), and rebooted. "Lo and behold", as Professor Tang used to say.
That was it. Kicking the portmapper, restarting the NFS services, re-exporting the shares, nothing I tried made a difference. Rebooting was just the easiest and most effective solution.
But it did cost siona 116 days of uptime for which we are all very sad :( Not a record, but still a good run. We'll miss you, 116 days uptime.
Monday, 18 June 2007
Sometimes, You Just Have to Reboot
Subscribe to: Post Comments (Atom)
For anyone who's had to cleanup some mail problems with Postfix configuration (or more often with other things, like anti-spam, tied in ...
In the course of troubleshooting the office Jabber server the other day, I came across some interesting info about the various caches that O...
For everyone who uses cron, you are familiar with the job schedule form: min hr day-of-month month day-of-week <command> A problem...
Post a Comment