At 01:34 AM 6/19/2007, Tony Nelson wrote:
>> Maybe do `lsof | grep __db` first?
Good morning, Tony.
I did that. There are four such files: __db.000 through __db.003. Of
those, only __db.000 is closed; the others are open. But, I can't find
anything in "ps aux" that looks like it might use rpm, not that I am
exactly up to speed on everything that might user rpm. I know that yum,
yum_updatesd, and probably pup and pirut, all use rpm, but don't know if
anything else does.
By the way, yum_updatesd is not running. I have it shut down because the
machine is located remotely (on the other side of the next city over from
here), I won't necessarily see notifications that updates are available,
and anyway I don't want to just blindly update everything; I'd rather
update only those things that need updating, mostly security-related stuff.
Should I try to run rpm -rebuilddb in the state the machine is in now, or
should I perhaps reboot first, and then delete those __db.00x files? I'll
need to drive over to the co-location facility to do that, but I guess I
can do that if need be.
>> When RPM seems to stop working, the locks are the most likely issue
Just to clarify, then, are the __db.00x files the lock files? I thought of
lock files but looked through /var/lock (where I would have expected such
files to be) and didn't see anything that was likely to be relevant.
