Gene Heskett wrote:
On Friday 20 June 2003 08:49, Jason Edgecombe wrote:
Look in /tmp/amanda/sendsize*debug on that system. It'll have
timestamped info on each estimate (probably level 0s and level 1s
for each fs). Figure out which one is taking so long. Any errors
in the system logs? What type of hardware RAID? What OS, and what
types of FSs?
Here are some snippets from the kernel messages:
--------------------------------------------------------------
SCSI subsystem driver Revision: 1.00
scsi0 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.10
<Adaptec AIC7902 Ultra320 SCSI adapter>
aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 101-133Mhz,
512 SCBs
scsi1 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 1.3.10
<Adaptec AIC7902 Ultra320 SCSI adapter>
aic7902: Ultra320 Wide Channel B, SCSI Id=7, PCI-X 101-133Mhz,
512 SCBs
...snip...
scsi2 : 3ware Storage Controller
Vendor: 3ware Model: Logical Disk 0 Rev: 1.0
Type: Direct-Access ANSI SCSI revision: 00
blk: queue c66caa18, I/O limit 4095Mb (mask 0xffffffff)
Vendor: 3ware Model: Logical Disk 1 Rev: 1.0
Type: Direct-Access ANSI SCSI revision: 00
blk: queue c66ca618, I/O limit 4095Mb (mask 0xffffffff)
...snip...
scsi3 : QLogic QLA2300 PCI to Fibre Channel Host Adapter: bus 3 device 1
irq 24
Firmware version: 3.01.13, Driver version 6.01.00-fo
Vendor: Tornado- Model: F2 V2.0 Rev:
Type: Direct-Access ANSI SCSI revision: 02
scsi(3:0:0:0): Enabled tagged queuing, queue depth 16.
Attached scsi disk sdc at scsi3, channel 0, id 0, lun 0
SCSI device sdc: 2109849600 512-byte hdwr sectors (1080243 MB)
sdc: sdc1
-------------------------------------
Software is Redhat linux 7.3 with custom redhat kernel 2.4.20-18.7bigmem
with the adaptec 79xx drivers from adaptec's web site.
The motherboard is a Supermicro X5DPi-G2 motherboard with dual Intel
2.4GHz Xeons and 4GB of RAM.
/dev/sda and /dev/sdb are 180GB ide drives attached to a 3ware ide to
scsi raid controller.
The Overland Neo2000 with LTO2 tape drive is on the Adaptec controller
and the hardware raid is on the Qlogic controller.
drives are as follows:
/dev/hda - 200GB ide disk - amanda holding disk
/dev/sda and /dev/sdab - 180GB each three partitions each; /boot and /
in a linux software RAID1 config. swap partitions on each disk
/dev/sdc - lvm volume - rebranded Axus Brownie BR-8000FC 2GB fibre
channel RAID5 array with IDE drives.
/home is a logical volume on the /dev/sdc volume group
all filesystems are ext3.
--- Volume group ---
VG Name vg01
VG Access read/write
VG Status available/resizable
VG # 0
MAX LV 256
Cur LV 1
Open LV 1
MAX LV Size 2 TB
Max PV 256
Cur PV 1
Act PV 1
VG Size 1005.94 GB
PE Size 64 MB
Total PE 16095
Alloc PE / Size 8800 / 550 GB
Free PE / Size 7295 / 455.94 GB
DLE's that are all on the same disk should all have the same 'spindle'
number. Its possible that amanda is running 2 or more sessions
against the same disk, in which case there will be some lost time due
to thrashing of the disk with uncoordinated seeks to different
partitions.
here is the pertinent part of the disklist:
localhost /boot root-tar 1
localhost / root-tar 1
localhost /home user-tar 2
Have you run an hdparm -Tt on the slow disks? On the raid?
# hdparm -tT /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 0.21 seconds =609.52 MB/sec
Timing buffered disk reads: 64 MB in 1.65 seconds = 38.79 MB/sec
# hdparm -tT /dev/sda
/dev/sda:
Timing buffer-cache reads: 128 MB in 0.21 seconds =609.52 MB/sec
Timing buffered disk reads: 64 MB in 2.07 seconds = 30.92 MB/sec
# hdparm -tT /dev/sdb
/dev/sdb:
Timing buffer-cache reads: 128 MB in 0.21 seconds =609.52 MB/sec
Timing buffered disk reads: 64 MB in 1.39 seconds = 46.04 MB/sec
# hdparm -tT /dev/sdc
/dev/sdc:
Timing buffer-cache reads: 128 MB in 0.23 seconds =556.52 MB/sec
Timing buffered disk reads: 64 MB in 1.27 seconds = 50.39 MB/sec
Not that I know of. amanda needs to figure out how much data there is
to be stored. This estimate phase is one of amanda's real strengths
as it makes for very efficient use of the available resources.
Thanks for the info.
#grep estimate sendsize.20030620014500.debug
sendsize[21320]: time 0.004: waiting for any estimate child
sendsize[21322]: estimate time for /boot level 0: 0.045
sendsize[21322]: estimate size for /boot level 0: 13010 KB
sendsize[21322]: estimate time for /boot level 1: 0.007
sendsize[21322]: estimate size for /boot level 1: 10 KB
sendsize[21320]: time 0.093: waiting for any estimate child
sendsize[21325]: estimate time for / level 0: 394.156
sendsize[21325]: estimate size for / level 0: 11711950 KB
sendsize[21325]: estimate time for / level 4: 103.619
sendsize[21325]: estimate size for / level 4: 551370 KB
sendsize[21320]: time 497.907: waiting for any estimate child
sendsize[21387]: estimate time for /home level 0: 2296.212
sendsize[21387]: estimate size for /home level 0: 12518750 KB
sendsize[21387]: estimate time for /home level 2: 1426.880
sendsize[21387]: estimate size for /home level 2: 909970 KB
sendsize[21387]: estimate time for /home level 3: 1514.321
sendsize[21387]: estimate size for /home level 3: 901020 KB
I'm going to make all of my disks in the disklist on spidle 1 and move
around the priority on my swap partitions.
Thanks,
Jason Edgecombe
|