RE: why too large ?
Perhaps in the next version, the AMANDA Powers-that-be could change this error message to read, The total size of all of today's dumps exceeds Tapelength size. Is more clear than, Dumps way too big, must skip incremental dumps. ...Unless my version is somehow untrue. wab -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Paul Bijnens Sent: Wednesday, November 19, 2003 1:29 PM To: Brian Cuttler Cc: [EMAIL PROTECTED]; Chris Knight Subject: Re: why too large ? Brian Cuttler wrote: So here is an interesting problem. Up until a couple weeks ago we had no problems at all running amdump on this system. Recently however its reporting that /maildb2 is too large for the tape. The confusing part to me though is that that is not the largest partition, the largest /maildb, which has about 50% greater disk occupancy, dumps without any problem at all. I believe amanda tries to tell you that it must skip the incremental dump, because there is no room on the tape to put both the full of /maildb and the incremental of /maildb2. What is your tapecapacity? I guess it's about 35 Gbyte. I notice too that a level 2 for /maildb is almost as large as a level 0. I guess the same is true for /maildb2. That means that the estimated size for /maildb2 with compression is about 15 Gbyte. 20+15 = 35Gbyte. This means that both of the filesystems do not fit on one tape anymore. When a filesystem is overdue, it gets a higher priority. When the priority gets high enough, you should observe the opposite: /maildb will fail, and /maildb2 will succeed. I'm not sure when that priority gets high enough, though. Is this what's happening? df -kl Filesystemkbytesused avail capacity Mounted on /dev/md/dsk/d10 7789512 2501372 521024533%/ /proc 0 0 0 0%/proc fd 0 0 0 0%/dev/fd mnttab 0 0 0 0%/etc/mnttab swap 7647720 8 7647712 1%/var/run swap 8580504 932792 764771211%/tmp /dev/md/dsk/d60 70555171 26346335 4350328538%/maildb2 /dev/md/dsk/d40 52421967 32849775 1904797364%/maildb /dev/dsk/c1t4d0s0352514988441 34890543 1%/amanda/work /dev/md/dsk/d30 17408538 6540222 1069423138%/export/home Any idea what amdump is choking on this partition ? thanks, Brian - Forwarded message from [EMAIL PROTECTED] - These dumps were to tape NOTES_DLT08. The next tape Amanda expects to use is: NOTES_DLT09. FAILURE AND STRANGE DUMP SUMMARY: wcnotes/maildb2 lev 1 FAILED [dumps way too big, must skip incremental dumps] STATISTICS: Total Full Daily Estimate Time (hrs:min)0:03 Run Time (hrs:min) 6:26 Dump Time (hrs:min)5:18 0:00 5:18 Output Size (meg) 20491.00.020491.0 Original Size (meg) 35425.20.035425.2 Avg Compressed Size (%)57.8--57.8 (level:#disks ...) Filesystems Dumped3 0 3 (1:2 2:1) Avg Dump Rate (k/s) 1100.1-- 1100.1 Tape Time (hrs:min)1:11 0:00 1:11 Tape Size (meg) 20491.10.020491.1 Tape Used (%) 62.10.0 62.1 (level:#disks ...) Filesystems Taped 3 0 3 (1:2 2:1) Avg Tp Write Rate (k/s) 4944.0-- 4944.0 NOTES: planner: Incremental of wcnotes:/maildb bumped to level 2. taper: tape NOTES_DLT08 kb 20982848 fm 3 [OK] DUMP SUMMARY: DUMPER STATS TAPER STATS HOSTNAME DISK L ORIG-KBOUT-KB COMP% MMM:SS KB/s MMM:SS KB/s --- - wcnotes / 1 927 128 13.8 1:141.7 0:02 64.4 wcnotes /export/home 1 4665087 1851744 39.7 32:03 962.8 6:13 4969.9 wcnotes /maildb2 31609440 19130880 60.5 284:35 1120.4 64:29 4944.7 wcnotes /maildb2 1 FAILED -- (brought to you by Amanda version 2.4.2p2) - End of forwarded message from [EMAIL PROTECTED] - -- Paul Bijnens, XplanationTel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, F6, * * quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH
Re: why too large ?
Hi, Brian Cuttler, on 19. November 2003 at 18:03 you wrote to amanda-users: BC So here is an interesting problem. BC Up until a couple weeks ago we had no problems at all running BC amdump on this system. Recently however its reporting that BC /maildb2 is too large for the tape. BC The confusing part to me though is that that is not the largest BC partition, the largest /maildb, which has about 50% greater disk BC occupancy, dumps without any problem at all. Do you use different dumptypes for the two directories? The compression settings could differ. btw, get to 2.4.4p1. It´s good. -- best regards, Stefan Stefan G. Weichinger mailto:[EMAIL PROTECTED]
Re: why too large ?
On Wed, 19 Nov 2003 at 12:03pm, Brian Cuttler wrote Up until a couple weeks ago we had no problems at all running amdump on this system. Recently however its reporting that /maildb2 is too large for the tape. Not exactly... FAILURE AND STRANGE DUMP SUMMARY: wcnotes/maildb2 lev 1 FAILED [dumps way too big, must skip incremental dumps] What it's telling you here is that the sum of *all the dumps* is too big. It had to FAIL one partition to get the tape usage below 'tapelength'. Look in the corresponding amdump.N file for all the gory details on how amanda made this decision. You may need to play with the bump* flags in amanda.conf, so that amanda is more willing to do a higher level dumps of filesystems rather than FAILing them. -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
Re: why too large ?
Brian Cuttler wrote: So here is an interesting problem. Up until a couple weeks ago we had no problems at all running amdump on this system. Recently however its reporting that /maildb2 is too large for the tape. The confusing part to me though is that that is not the largest partition, the largest /maildb, which has about 50% greater disk occupancy, dumps without any problem at all. I believe amanda tries to tell you that it must skip the incremental dump, because there is no room on the tape to put both the full of /maildb and the incremental of /maildb2. What is your tapecapacity? I guess it's about 35 Gbyte. I notice too that a level 2 for /maildb is almost as large as a level 0. I guess the same is true for /maildb2. That means that the estimated size for /maildb2 with compression is about 15 Gbyte. 20+15 = 35Gbyte. This means that both of the filesystems do not fit on one tape anymore. When a filesystem is overdue, it gets a higher priority. When the priority gets high enough, you should observe the opposite: /maildb will fail, and /maildb2 will succeed. I'm not sure when that priority gets high enough, though. Is this what's happening? df -kl Filesystemkbytesused avail capacity Mounted on /dev/md/dsk/d10 7789512 2501372 521024533%/ /proc 0 0 0 0%/proc fd 0 0 0 0%/dev/fd mnttab 0 0 0 0%/etc/mnttab swap 7647720 8 7647712 1%/var/run swap 8580504 932792 764771211%/tmp /dev/md/dsk/d60 70555171 26346335 4350328538%/maildb2 /dev/md/dsk/d40 52421967 32849775 1904797364%/maildb /dev/dsk/c1t4d0s0352514988441 34890543 1%/amanda/work /dev/md/dsk/d30 17408538 6540222 1069423138%/export/home Any idea what amdump is choking on this partition ? thanks, Brian - Forwarded message from [EMAIL PROTECTED] - These dumps were to tape NOTES_DLT08. The next tape Amanda expects to use is: NOTES_DLT09. FAILURE AND STRANGE DUMP SUMMARY: wcnotes/maildb2 lev 1 FAILED [dumps way too big, must skip incremental dumps] STATISTICS: Total Full Daily Estimate Time (hrs:min)0:03 Run Time (hrs:min) 6:26 Dump Time (hrs:min)5:18 0:00 5:18 Output Size (meg) 20491.00.020491.0 Original Size (meg) 35425.20.035425.2 Avg Compressed Size (%)57.8--57.8 (level:#disks ...) Filesystems Dumped3 0 3 (1:2 2:1) Avg Dump Rate (k/s) 1100.1-- 1100.1 Tape Time (hrs:min)1:11 0:00 1:11 Tape Size (meg) 20491.10.020491.1 Tape Used (%) 62.10.0 62.1 (level:#disks ...) Filesystems Taped 3 0 3 (1:2 2:1) Avg Tp Write Rate (k/s) 4944.0-- 4944.0 NOTES: planner: Incremental of wcnotes:/maildb bumped to level 2. taper: tape NOTES_DLT08 kb 20982848 fm 3 [OK] DUMP SUMMARY: DUMPER STATSTAPER STATS HOSTNAME DISK L ORIG-KBOUT-KB COMP% MMM:SS KB/s MMM:SS KB/s --- - wcnotes / 1 927 128 13.8 1:141.7 0:02 64.4 wcnotes /export/home 1 4665087 1851744 39.7 32:03 962.8 6:13 4969.9 wcnotes /maildb2 31609440 19130880 60.5 284:35 1120.4 64:29 4944.7 wcnotes /maildb2 1 FAILED -- (brought to you by Amanda version 2.4.2p2) - End of forwarded message from [EMAIL PROTECTED] - -- Paul Bijnens, XplanationTel +32 16 397.511 Technologielaan 21 bus 2, B-3001 Leuven, BELGIUMFax +32 16 397.512 http://www.xplanation.com/ email: [EMAIL PROTECTED] *** * I think I've got the hang of it now: exit, ^D, ^C, ^\, ^Z, ^Q, F6, * * quit, ZZ, :q, :q!, M-Z, ^X^C, logoff, logout, close, bye, /bye, * * stop, end, F3, ~., ^]c, +++ ATH, disconnect, halt, abort, hangup, * * PF4, F20, ^X^X, :D::D, KJOB, F14-f-e, F8-e, kill -1 $$, shutdown, * * kill -9 1, Alt-F4, Ctrl-Alt-Del, AltGr-NumLock, Stop-A, ...* * ... Are you sure? ... YES ... Phew ... I'm out * ***