siis_timeout with port multiplier on 9.0R

[Matthew Gamble]

We have a box with 3 SiI3124 SATA controllers and 9 CFI-B53PM 5 Port Backplane port multipliers (the "backblaze storage pod").  Under intense IO (ZFS rebuild, presently) the system will lock up all IO for 3-4 minutes and the following entry appears in the dmesg:

siisch11: Timeout on slot 30
siisch11: siis_timeout is 00040000 ss 65000000 rs 65000000 es 00000000 sts 80192000 serr 00000000
siisch11:  ... waiting for slots 25000000
siisch11: Timeout on slot 26
siisch11: siis_timeout is 00040000 ss 65000000 rs 65000000 es 00000000 sts 80192000 serr 00000000
siisch11:  ... waiting for slots 21000000
siisch11: Timeout on slot 29
siisch11: siis_timeout is 00040000 ss 65000000 rs 65000000 es 00000000 sts 80192000 serr 00000000
siisch11:  ... waiting for slots 01000000
siisch11: Timeout on slot 24
siisch11: siis_timeout is 00040000 ss 65000000 rs 65000000 es 00000000 sts 80192000 serr 00000000

The errors are on different siisch devices so its not likely to be a SATA cable issue unless multiple cables all went bad at the same time.  On the advice of some other posts to the mailing list I've already tried locking the SATA rev to one with the following in /boot/loader.conf which didn't

hint.siisch.0.sata_rev=1
hint.siisch.1.sata_rev=1
hint.siisch.2.sata_rev=1
hint.siisch.3.sata_rev=1
hint.siisch.4.sata_rev=1
hint.siisch.5.sata_rev=1
hint.siisch.6.sata_rev=1
hint.siisch.7.sata_rev=1
hint.siisch.8.sata_rev=1
hint.siisch.9.sata_rev=1
hint.siisch.10.sata_rev=1
hint.siisch.11.sata_rev=1

From time to time this is also causing one of the attached drives to go offline:

siisch0: siis_timeout is 00040000 ss 40000000 rs 40000000 es 00000000 sts 801f2000 serr 00000000
(ada0:siisch0:0:0:0): lost device
(ada0:siisch0:0:0:0): removing device entry
ada0 at siisch0 bus 0 scbus0 target 0 lun 0
ada0: <WDC WD30EZRX-00MMMB0 80.00A80> ATA-8 SATA 3.x device
ada0: 150.000MB/s transfers (SATA 1.x, UDMA6, PIO 8192bytes)
ada0: Command Queueing enabled
ada0: 2861588MB (5860533168 512 byte sectors: 16H 63S/T 16383C)
ada0: Previously was known as ad4
siisch11: Timeout on slot 30

When the drive goes offline that causes the ZFS rebuild to restart, and so it's never finishing the rebuild of the array.  Does anyone have any insight into what could be causing the timeouts and what we can do to resolve them?  Right now my priority is to get the system a bit more stable so the current ZFS rebuild can complete – right now it's been doing the same rebuild for just over 6 days and the timeouts and drive drop offs are causing it to restart constantly.





________________________________

 This electronic message contains information from Primus Telecommunications Canada Inc. ("PRIMUS") , which may be legally privileged and confidential. The information is intended to be for the use of the individual(s) or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this electronic message in error, please notify us by telephone or e-mail (to the number or address above) immediately. Any views, opinions or advice expressed in this electronic message are not necessarily the views, opinions or advice of PRIMUS. It is the responsibility of the recipient to ensure that any attachments are virus free and PRIMUS bears no responsibility for any loss or damage arising in any way from the use thereof.The term "PRIMUS" includes its affiliates.

________________________________
 Pour la version en français de ce message, veuillez voir
http://www.primustel.ca/fr/legal/cs.htm