On Sunday 06 October 2002 11:02, Gene Heskett wrote: >On Sunday 06 October 2002 10:23, Gene Heskett wrote:
This may now be a case of the little boy who cryed wolf. 8-( I just discoverd that this early 10/05 date also corresonds to my adding a "-nice 5" to the end of the crontab entry. 2.2.5-40 has now been re-installed, amanda rebuilt and re-installed, and a fresh amdump is now running. I took that extra -nice 5 out of the crontab because it really didn't seem to free up the machine when amanda is running. If it runs ok in the morning, I'll add that to the crontab entry again for the next test. >>Greets list; >> >>A heads up on what looks to be a major problem in that the last >>update/security fix to glibc seems to have somehow killed amanda, >>as follows. >> >>I installed this latest glibc-2.2.5-40 set of packages about an >> hour after they showed up on updates.redhat.com. It was late >> enough that when I did the install, amanda had already started. >> >>The next morning (Oct 4) all I had in my amdump report were >> RESULTS MISSING messages for all 37 entries in the disklist. I >> blamed that in the fact that I had probably changed something in >> the middle of the run that fubared amanda. >> >>This was repeated this morning also. So I re-installed the >>glibc-2.2.5-39 package set, rebooted twice since setiathome >> didn't restart the first time, and I've got a shell open as >> amanda running a fresh amdump to see if it will now work. >> >>For me and this machine, reverting isn't a *serious* problem as >> I've a good firewall between this machine and the net. But for >> those who are tryng to backup a machine thats also your >> firewall, this could be a problem. >> >>I will post a fail/success report when I know if this single >>reversion fixes the problem. And then if it does, I'll >> re-install 2.2.5-40, and rebuild/re-install 2.4.3b4-20020930 >> amanda to see if that fixes it. This is the first time I've >> seen a security fix break amanda if indeed this is the case. > >Further update, according to amstatus, it has now proceeded past > the estimate phase and is actually writing data to the tape, > ditto according to the access led on the drive. So it appears > that the reversion to glibc-2.2.5-39 did indeed work. As to > testing the 2.2.5-40 version against a rebuilt amanda, that will > take a few more hours. Stay tuned as they say on the radio. :-) -- Cheers, Gene AMD K6-III@500mhz 320M Athlon1600XP@1400mhz 512M 99.17% setiathome rank, not too shabby for a WV hillbilly
