Author: Eric Siegerman <erics AT telepres DOT com>
Date: Wed, 2 Jun 2004 15:02:49 -0400
That's pretty much what we do for our weekly level-0's-to-tape backups. We simply avoid inserting a tape on Friday night, which is when that configuration runs. Amdump goes into degraded mode; I set
I thought so too. But when dump is running from one disk to the other, it reports speeds about 20% higher (~10MB/s) than it does when running from disk to tape (~8MB/s). I can get dump to stream by s
Author: Gene Heskett <gene.heskett AT verizon DOT net>
Date: Wed, 2 Jun 2004 16:12:20 -0400
I get the impression that you are backing up something thats on the same drive as the holding disk, and that possibly dma may also be missfireing somehow. Any current ide drive can do 30+ Mb/sec if l
The speed of the dump or tar programs -> tape should be totally irrelavant since amanda never does that if the holding disk is in use. I don't think they go straight to tape even when the holding dis
Is that just a burst out of the cache, or can they read dis-contiguous files, seek around to other files, wait for latency, and write all at the same time that fast? Or even half that fast? If so, an
Author: "Stefan G. Weichinger" <monitor AT oops.co DOT at>
Date: Thu, 3 Jun 2004 00:37:59 +0200
Hi, Glenn, on Donnerstag, 03. Juni 2004 at 00:21 you wrote to amanda-users: Just look for the parameter dumporder in the example for the amanda.conf file. It's located in the example-subdir of your a
On Solaris x86, one thing that terribly degrades IDE disk performance is if the driver does not use dma but uses pio mode. There are even reports of diagnostic tools saying the drive is using dma, bu
Author: "Stefan G. Weichinger" <monitor AT oops.co DOT at>
Date: Thu, 3 Jun 2004 01:00:50 +0200
Hi, Jon, on Donnerstag, 03. Juni 2004 at 00:48 you wrote to amanda-users: OT: hdparm may help. Try hdparm -i /dev/hdX (substitute X with the fitting char for your prim/sec Master/Slave whatever ....)
--On Wednesday, June 02, 2004 18:48:46 -0400 Jon LaBadie <jon AT jgcomp DOT If it's linux, try using hdparm to verify the modes and speed of your disk. Like Jon says, a good drive can have terrible p
Author: Gene Heskett <gene.heskett AT verizon DOT net>
Date: Wed, 2 Jun 2004 21:56:52 -0400
It's true of any os, not just solaris, and I thought of that, but got lost and didn't mention it in my novel on this. Thanks for reminding me. I think I may have assumed he had all that checked out.
Author: Gene Heskett <gene.heskett AT verizon DOT net>
Date: Wed, 2 Jun 2004 21:52:55 -0400
Well, in fairness, thats the hdparm -tT rateings I'm quoting, which is generally a 1 or 2 second burst, either from the cache, or from the surface itself. This does NOT take into consideration seek t
Author: Gene Heskett <gene.heskett AT verizon DOT net>
Date: Wed, 2 Jun 2004 22:00:28 -0400
And then to measure how fast that drive can go, do "hdparm -tT /dev/hdX" It should look something like this: [root@coyote root]# hdparm -tT /dev/hda /dev/hda: Timing buffer-cache reads: 920 MB in 2.0
Gene's mention of compression reminded me. If you are using disk drive compression you have to feed data at a higher rate to the drive than if it is already compressed with gzip and then sent to the
hdparm is a nifty addition to my system monitoring toolkit (top, gnome's cpu monitor icon, sticking my ear close to a drive to listen for seeks, eyeballing the disk activity LED, and dump's data tran
Whoops, thanks for the correction. -- Jon H. LaBadie jon AT jgcomp DOT com JG Computing 4455 Province Line Road (609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
Author: "Stefan G. Weichinger" <monitor AT oops.co DOT at>
Date: Thu, 3 Jun 2004 09:12:13 +0200
Hi, Glenn, on Donnerstag, 03. Juni 2004 at 06:51 you wrote to amanda-users: pio looks different ;-) Would be much slower. Getting away from AMANDA-topics: You should get your IDE-setup straight. A ma
Author: "Stefan G. Weichinger" <monitor AT oops.co DOT at>
Date: Thu, 3 Jun 2004 09:26:55 +0200
Hi, Stefan G. Weichinger, on Donnerstag, 03. Juni 2004 at 09:12 you wrote to amanda-users: I forgot: To use udma5 you have to use an appropriate ATA (80 core) cable for the drive. -- best regards, St
Using many holding disks helps here; I have about twenty disks, many of them old 18GB drives. Later this year I will upgrade the library from SDLT220 to SDLT600, and these drives will eat 36MBytes/se
Author: Martin Hepworth <martinh AT solid-state-logic DOT com>
Date: Thu, 03 Jun 2004 09:02:12 +0100
Glen I'd check the logic in the kernel code that checks for 80 way cables etc etc. When I last looked at 2.4.19 it was doing some weird stuff - do the checks then reset the flag where the bit table f