[OmniOS-discuss] [SPAM] Bizarre zfs-related hang in omnios r151008 on 1-CPU VM

[Saso Kiselkov]

On 12/6/13, 5:39 AM, Rob Logan wrote:
> 
>> on the latest stable omnios release. When I'm running in VMware Fusion
>> on a 1-CPU VM and doing any significant write IO to the pool (e.g. just
>>    atomic_dec_32_nv+8()
>>    dbuf_read+0x179(ffffff00d2393600, ffffff00c72f98f0, a)
>>    dmu_tx_check_ioerr+0x76(ffffff00c72f98f0, ffffff00d2279cf0, 0, 1e0)
>>    dmu_tx_count_write+0x395(ffffff00ce0536e0, 3c04000, 4000)
>>    dmu_tx_hold_write+0x5a(ffffff00d1a55300, 4009, 3c04000, 4000)
>>    zfs_write+0x3e3(ffffff00d09ef540, ffffff00028e7e60, 0,
>> ffffff00cd511748, 0)
>>    fop_write+0x5b(ffffff00d09ef540, ffffff00028e7e60, 0,
>> ffffff00cd511748, 0)
>>    write+0x250(1, 440660, 4000)
>>    sys_syscall+0x17a()
> 
> 
> doing the normal re-write of root in r151008 three times 
> into lz4 didn’t have any issues on my 2cpu 2G vbox
> 
> root at OmniOS:~# lspci
> 00:02.0 VGA compatible controller: InnoTek Systemberatung GmbH VirtualBox Graphics Adapter
> 00:04.0 System peripheral: InnoTek Systemberatung GmbH VirtualBox Guest Service
> 00:05.0 Multimedia audio controller: Intel Corporation 82801AA AC'97 Audio Controller (rev 01)
> 00:07.0 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08)
> 00:11.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 02)
> 00:18.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev f2)
> 00:19.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev f2)
> 00:1f.0 ISA bridge: Intel Corporation 82801GBM (ICH7-M) LPC Interface Bridge (rev 02)
> 00:1f.2 SATA controller: Intel Corporation 82801HM/HEM (ICH8M/ICH8M-E) SATA Controller [AHCI mode] (rev 02)
> 00:1f.4 USB controller: Apple Inc. KeyLargo/Intrepid USB
> 
> root at OmniOS:~# zfs get all | grep refcompressratio
> rpool                                refcompressratio              1.00x                                 -
> rpool/ROOT                           refcompressratio              1.00x                                 -
> rpool/ROOT/start                     refcompressratio              1.85x                                 -
> rpool/ROOT/work                      refcompressratio              1.98x                                 -
> rpool/ROOT/work at 2013-12-05-19:24:16  refcompressratio              1.85x                                 -
> 
> not sure how to reproduce.

You need a 1-CPU system. As Matt pointed out, the hang is most probably
caused by a deadlock that was resolved in e722410.

OmniTI: I believe rolling this into the next weekly patch cycle might be
kind of important?

Cheers,
-- 
Saso