[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] Kernel 3.7.[12] - irq 16: nobody cared



On 16/01/2013 9:05 PM, Jan Beulich wrote:
On 16.01.13 at 10:54, Steven Haigh <netwiz@xxxxxxxxx> wrote:
So far, I have:
# uptime
   20:50:40 up 1 day,  1:11,  1 user,  load average: 0.36, 0.17, 0.13

As I mentioned, I moved the sata card to the second 16x PCIe slot in the
mainboard - which changed the IRQ from 16 to 19. Currently I see:
# grep sata_mv /proc/interrupts
   19:   21243495  xen-pirq-ioapic-level  sata_mv

Which is interestingly more than the onboard SATA ports:
# grep ahci /proc/interrupts
   50:    9004117  xen-pirq-msi       ahci
Whether the former count is too high depends on the I/O amount
going through each controller. Of course it is possible for there to
be spikes that usually don't reach the 99,900 cutoff point, but
once in a while do. Figuring whether that's the case would require
adding a little bit more verbosity to
kernel/irq/spurious.c:note_interrupt(), e.g. to warn when having
reached half the threshold.

Interestingly, I just realised I have 3 of the 4 drives in this RAID6 on the sata_mv card. I did originally think I had 2 drives on the onboard SATA ports, and the other 2 on the sata_mv card. This would mean 3/4 of the IO would be going via this card - but only 1/4 on the onboard.

# lsdrv
PCI [ahci] 00:1f.2 SATA controller: Intel Corporation 6 Series/C200 Series Chipset Family SATA AHCI Controller (rev 05)
.scsi 0:0:0:0 ATA ST380815AS {6RAB72DZ}
..sda 74.53g [8:0] Partitioned (dos)
. .sda1 200.00m [8:1] MD raid1 (1/2) (w/ sdb1) in_sync 'localhost.localdomain:0' {9f19116a-d280-8216-cc87-af34eae68242}
. ..md0 199.99m [9:0] MD v1.0 raid1 (2) clean
. . . Partitioned (dos) {6578dbc0-9e07-4ccc-8eff-15f2a1da8df1}
. . .Mounted as /dev/md0 @ /boot
. .sda2 74.33g [8:2] MD raid1 (1/2) (w/ sdb2) in_sync 'localhost.localdomain:1' {afb92c19-b9b1-e3ae-07af-315d738e38be}
.  .md1 74.33g [9:1] MD v1.1 raid1 (2) clean
. . PV LVM2_member 74.33g used, 0 free {2koqPs-U1IA-9erV-ua4N-mxW1-BhRs-V3mlAH}
.   .VG RAID1 74.33g 0 free {HEGjco-Ptil-M5ZG-2qQR-zNo4-3cc5-b9Z3Kj}
. .dm-0 9.77g [253:0] LV xenhost ext4 {d2fa50d5-1a51-4599-9b72-f38f86b8f99e}
.    ..Mounted as /dev/mapper/RAID1-xenhost @ /
. .dm-7 64.56g [253:7] LV zeus.vm ext4 {67310780-b15c-47e4-812e-d954aa7d8e3b}
.scsi 1:0:0:0 ATA ST380815AS {6QZ6L9SD}
..sdb 74.53g [8:16] Partitioned (dos)
. .sdb1 200.00m [8:17] MD raid1 (0/2) (w/ sda1) in_sync 'localhost.localdomain:0' {9f19116a-d280-8216-cc87-af34eae68242}
. ..md0 199.99m [9:0] MD v1.0 raid1 (2) clean
. . Partitioned (dos) {6578dbc0-9e07-4ccc-8eff-15f2a1da8df1} . .sdb2 74.33g [8:18] MD raid1 (0/2) (w/ sda2) in_sync 'localhost.localdomain:1' {afb92c19-b9b1-e3ae-07af-315d738e38be}
.  .md1 74.33g [9:1] MD v1.1 raid1 (2) clean
. PV LVM2_member 74.33g used, 0 free {2koqPs-U1IA-9erV-ua4N-mxW1-BhRs-V3mlAH}
.scsi 2:x:x:x [Empty]
.scsi 3:0:0:0 ATA ST2000VX000-9YW1 {Z1E10QQJ}
..sdc 1.82t [8:32] MD raid6 (3/4) (w/ sdd,sde,sdf) in_sync 'xenhost.lan.crc.id.au:2' {cd8cc032-4898-fa88-3ba1-af64cf91583b}
. .md2 3.64t [9:2] MD v1.2 raid6,left-sym (4) active, 128k Chunk
. . PV LVM2_member 2.12t used, 1.52t free {8pyp2G-D268-fqKW-mBvf-wZbI-Qurt-aeTvOh}
.  .VG vg_raid6 3.64t 1.52t free {UrqTRc-AozJ-2RDf-qcZB-UdX3-tno9-3KHjjv}
. .dm-6 2.00t [253:6] LV fileshare xfs {af405459-7569-4d82-82d9-ca27912316c7} . .dm-3 10.00g [253:3] LV lamp.vm ext4 {67310780-b15c-47e4-812e-d954aa7d8e3b} . .dm-2 40.00g [253:2] LV mail.vm ext4 {67310780-b15c-47e4-812e-d954aa7d8e3b}
.   .dm-4 20.00g [253:4] LV remotedesktop.vm Partitioned (dos)
. .dm-5 2.00g [253:5] LV template.vm ext4 {67310780-b15c-47e4-812e-d954aa7d8e3b} . .dm-1 50.00g [253:1] LV tsm.vm ext4 {67310780-b15c-47e4-812e-d954aa7d8e3b}
.scsi 4:x:x:x [Empty]
.scsi 5:x:x:x [Empty]
PCI [sata_mv] 04:00.0 SCSI storage controller: Marvell Technology Group Ltd. 88SX7042 PCI-e 4-port SATA-II (rev 02)
.scsi 6:0:0:0 ATA ST2000VX000-9YW1 {Z1E11E7R}

..sdd 1.82t [8:48] MD raid6 (0/4) (w/ sdc,sde,sdf) in_sync 'xenhost.lan.crc.id.au:2' {cd8cc032-4898-fa88-3ba1-af64cf91583b}
. .md2 3.64t [9:2] MD v1.2 raid6,left-sym (4) active, 128k Chunk
. PV LVM2_member 2.12t used, 1.52t free {8pyp2G-D268-fqKW-mBvf-wZbI-Qurt-aeTvOh}
.scsi 7:x:x:x [Empty]
.scsi 8:0:0:0 ATA ST2000VX000-9YW1 {Z1E0MD58}
..sde 1.82t [8:64] MD raid6 (1/4) (w/ sdc,sdd,sdf) in_sync 'xenhost.lan.crc.id.au:2' {cd8cc032-4898-fa88-3ba1-af64cf91583b}
. .md2 3.64t [9:2] MD v1.2 raid6,left-sym (4) active, 128k Chunk
. PV LVM2_member 2.12t used, 1.52t free {8pyp2G-D268-fqKW-mBvf-wZbI-Qurt-aeTvOh}
.scsi 9:0:0:0 ATA ST2000VX000-9YW1 {Z1E17C3X}
.sdf 1.82t [8:80] MD raid6 (2/4) (w/ sdc,sdd,sde) in_sync 'xenhost.lan.crc.id.au:2' {cd8cc032-4898-fa88-3ba1-af64cf91583b}
  .md2 3.64t [9:2] MD v1.2 raid6,left-sym (4) active, 128k Chunk
PV LVM2_member 2.12t used, 1.52t free {8pyp2G-D268-fqKW-mBvf-wZbI-Qurt-aeTvOh}

I'm going to leave it as is at the moment to see if it happens again as it has been randomly over the last 3-4 weeks. I'll try to pull any info off this time before rebooting the system - as I only recently found this problem. Hopefully, either changing the slot, or even just reseating the card may have had some effect - but I guess only time will tell.

--
Steven Haigh

Email: netwiz@xxxxxxxxx
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299



Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.