Hi Dan, On Fri, 2015-04-03 at 11:41 -0400, Dan Lane wrote: > I recently rebuilt my storage server (again) using Fedora 21 Server > (64-bit). Once everything was up and running I built out my target > configuration and confirmed it was working. This was using the > original Kernel 3.17.4. > > At this point I decided to update my update my server. Once I was > done with all the updates, I found that target now refuses to start. > Running "journalctl -xe" shows the following errors: > > Apr 03 02:24:36 labsan1.home.lan kernel: Rounding down aligned > max_sectors from 4294967295 to 4294967288 > Apr 03 02:24:36 labsan1.home.lan kernel: dev[ffff8800c8ffa000]: Passed > optimal_sectors 8192 cannot be greater than hw_max_sectors: 512 > Apr 03 02:24:36 labsan1.home.lan kernel: qla2xxx > [0000:08:01.1]-00af:6: Performing ISP error recovery - > ha=ffff8800c7098000. > Apr 03 02:24:37 labsan1.home.lan kernel: qla2xxx > [0000:08:01.1]-500a:6: LOOP UP detected (4 Gbps). > Apr 03 02:24:37 labsan1.home.lan kernel: qla2xxx > [0000:08:01.0]-00af:5: Performing ISP error recovery - > ha=ffff8800c79f4000. > Apr 03 02:24:38 labsan1.home.lan kernel: qla2xxx > [0000:08:01.0]-500a:5: LOOP UP detected (4 Gbps). > Apr 03 02:24:38 labsan1.home.lan kernel: qla2xxx > [0000:0d:01.1]-00af:4: Performing ISP error recovery - > ha=ffff8800c748c000. > Apr 03 02:24:39 labsan1.home.lan kernel: qla2xxx > [0000:0d:01.1]-500a:4: LOOP UP detected (4 Gbps). > Apr 03 02:24:39 labsan1.home.lan kernel: qla2xxx > [0000:0d:01.0]-00af:3: Performing ISP error recovery - > ha=ffff8800359ac000. > Apr 03 02:24:40 labsan1.home.lan kernel: qla2xxx > [0000:0d:01.0]-500a:3: LOOP UP detected (4 Gbps). > Apr 03 02:24:40 labsan1.home.lan target[1149]: Restore failed, 2 errors: > Apr 03 02:24:40 labsan1.home.lan target[1149]: Storage Object > block/data: Cannot set attribute optimal_sectors: [Errno 22] Invalid > argument, skipped > Apr 03 02:24:40 labsan1.home.lan target[1149]: Storage Object > block/data: Cannot find attribute: fabric_max_sectors, skipped > Apr 03 02:24:40 labsan1.home.lan systemd[1]: target.service: main > process exited, code=exited, status=255/n/a > Apr 03 02:24:40 labsan1.home.lan systemd[1]: Failed to start Restore > LIO kernel target configuration. > -- Subject: Unit target.service has failed > -- Defined-By: systemd > -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel > -- > -- Unit target.service has failed. > -- > -- The result is failed. > > I found a few other results on google that seemed to mostly match > this, but they talk about this being patched already. > > Package versions: > python-rtslib-2.1.fb52-1.fc21.noarch > targetcli-2.1.fb39-1.fc21.noarch > > uname -a: > Linux labsan1.home.lan 3.19.3-200.fc21.x86_64 #1 SMP Thu Mar 26 > 21:39:42 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux > This looks like a -fb specific bug. Even if setting an attribute fails, it should not abort bringing up the device. Adding Andy CC'. A simple work-around is to change the saved optimal_sectors value for the backend device in question from 8192 to 512 to match hw_max_sectors. --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
