On Jul 19 09:25:22, h...@stare.cz wrote: > On Jul 18 21:53:02, h...@stare.cz wrote: > > Below is what my daily (nightly) dump said. In short, > > > > read error from /dev/rsd5a: Invalid argument: [block -60912738498]: > > count=10240 > > > > Thats daily.loca running `dump -0 -a -l 0 -f - /home > /backup/dump.home' > > The filesystem (/dev/sd5a = /home) is fine; fsck says it's clean. > > > > What could be causing this? During the dump, there was some activity > > on the filesystem (bitcoind catching up to the blockchain). > > Same thing on another machine: > > errors dumping /home: > DUMP: Date of this level 0 dump: Wed Jul 19 01:30:48 2017 > DUMP: Date of last level 0 dump: the epoch > DUMP: Dumping /dev/rsd2a (/home) to standard output > DUMP: mapping (Pass I) [regular files] > DUMP: mapping (Pass II) [directories] > DUMP: estimated 105512095 tape blocks. > DUMP: Volume 1 started at: Wed Jul 19 01:31:34 2017 > DUMP: dumping (Pass III) [directories] > DUMP: dumping (Pass IV) [regular files] > DUMP: 0.60% done, finished in 13:54 > DUMP: 1.13% done, finished in 14:36 > DUMP: 1.70% done, finished in 14:27 > DUMP: 2.17% done, finished in 15:01 > DUMP: 2.71% done, finished in 14:56 > DUMP: 3.22% done, finished in 15:00 > DUMP: 3.78% done, finished in 14:50 > DUMP: 4.30% done, finished in 14:50 > DUMP: 4.81% done, finished in 14:50 > DUMP: 5.31% done, finished in 14:50 > DUMP: 5.86% done, finished in 14:44 > DUMP: 6.41% done, finished in 14:35 > DUMP: 6.91% done, finished in 14:35 > DUMP: 7.44% done, finished in 14:31 > DUMP: 7.97% done, finished in 14:26 > DUMP: 8.53% done, finished in 14:17 > DUMP: 9.06% done, finished in 14:13 > DUMP: 9.62% done, finished in 14:05 > DUMP: 10.17% done, finished in 13:58 > DUMP: 10.84% done, finished in 13:42 > DUMP: 11.47% done, finished in 13:30 > DUMP: 12.10% done, finished in 13:19 > DUMP: 12.73% done, finished in 13:08 > DUMP: 13.49% done, finished in 12:49 > DUMP: 14.32% done, finished in 12:28 > DUMP: 15.13% done, finished in 12:09 > DUMP: 16.00% done, finished in 11:48 > DUMP: 17.08% done, finished in 11:19 > DUMP: 18.28% done, finished in 10:48 > DUMP: 19.82% done, finished in 10:06 > DUMP: 23.18% done, finished in 8:33 > DUMP: 28.19% done, finished in 6:47 > DUMP: 33.37% done, finished in 5:29 > DUMP: 40.13% done, finished in 4:13 > DUMP: 47.48% done, finished in 3:13 > DUMP: 55.27% done, finished in 2:25 > DUMP: DUMP: DUMP: read error from /dev/rsd2a: Invalid argument: [block > -57913999616]: count=7168 > read error from /dev/rsd2a: Invalid argument: [block -57913999602]: > count=10240 > DUMP: read error from /dev/rsd2a: Invalid argument: [block -57913999616]: > count=512 > DUMP: read error from /dev/rsd2a: Invalid argument: [block -57913999582]: > count=10240 > DUMP: read error from /dev/rsd2a: Invalid argument: [block -57913999602]: > count=512 > read error from /dev/rsd2a: Invalid argument: [block -57913999615]: count=512 > DUMP: DUMP: read error from /dev/rsd2a: Invalid argument: [block > -57913999582]: count=512 [...]
On Jul 20 11:46:16, skin...@britvault.co.uk wrote: > On Tue, 18 Jul 2017 21:53:03 +0200 Jan Stary wrote: > > ... During the dump, there was some activity on the filesystem ... > > Try altering your scripts stop relevant daemons, then umount the > partition before dumping & see if the problem persists. If nothing is happening on the filesystem, it dumps OK; it does not matter that it's mounted. > See sections '5.0 Conclusions' and '2.0 Active Tests' of Elizabeth D. > Zwicky's paper "Torture-testing Backup and Archive Programs: Things You > Ought to Know But Probably Would Rather Not". > > http://www.CoreDumps.De/doc/dump/zwicky/testdump.doc.html > > Specifically for /home;- also see '/etc/nologin' in login(1), which file > I've found to be usefully created & nuked by my dump driving scripts. In my case, it does not make a difference that someone is logged in, but that there are files being written (heavily) during the dump. At any rate, the negative block numbers seem wrong. Jan
