Bug 848942

This solution works but is not ideal. Opened an upstream bug at https://bugs.freedesktop.org/show_bug.cgi?id=55431 . In the meantime I guess we'll do your proposed fix.

We are still seeing this on F18 with scsi-target-utils-1.0.32-2.fc18.x86_64.

Jan 31 09:54:19 xxx systemd[1]: Starting tgtd iSCSI target daemon...
Jan 31 09:54:19 xxx tgtd[698]: librdmacm: Warning: couldn't read ABI version.
Jan 31 09:54:19 xxx tgtd[698]: librdmacm: Warning: assuming: 4
Jan 31 09:54:19 xxx tgtd[698]: librdmacm: Fatal: unable to get RDMA device list
Jan 31 09:54:19 xxx tgtd[698]: tgtd: iser_ib_init(3376) Failed to initialize RDMA; load kernel modules?
Jan 31 09:54:19 xxx tgtadm[700]: tgtadm: failed to send request hdr to tgt daemon, Transport endpoint is not connected
Jan 31 09:54:19 xxx tgtd[698]: tgtd: work_timer_start(146) use timer_fd based scheduler
Jan 31 09:54:19 xxx tgtd[698]: tgtd: bs_init(313) use signalfd notification
Jan 31 09:54:19 xxx systemd[1]: tgtd.service: control process exited, code=exited status=107
Jan 31 09:54:21 xxx tgtd[698]: tgtd: iscsi/iser.c:3429: iser_ib_release: Assertion `list_empty(&iser_conn_list)' failed.
Jan 31 09:54:21 xxx systemd[1]: tgtd.service: main process exited, code=killed, status=6/ABRT
Jan 31 09:54:21 xxx systemd[1]: Failed to start tgtd iSCSI target daemon.
Jan 31 09:54:21 xxx systemd[1]: Unit tgtd.service entered failed state

As mentioned above it looks as if the tgtadm commands are trying to run before tgtd is ready.

The sleep 5 fix seems to work but needs to have the path explicitly included, ie
ExecStartPost=/usr/bin/sleep 5

Is there a possibility to get the suggested sleep fix into fedora packages while we are waiting for progress on the upstream bug?

Andy, Comment on the upstream ticket says this:

Could you paste your unit file?

Maybe the best would be to split the ExecStartPost out into its own unit an simply order that after your main unit?

----

We require tgtd for OpenStack Cinder and this is currently a show stopper on Fedora 17/18. I see no reason not to package the work around ('ExecStartPost=/usr/bin/sleep 5') while we wait on the upstream to get fixed (and released). It has just been too long...

Sorry for the delay. Yes, we'll do the workaround right away for F17/18.

scsi-target-utils-1.0.24-7.fc17 has been submitted as an update for Fedora 17.
https://admin.fedoraproject.org/updates/scsi-target-utils-1.0.24-7.fc17

scsi-target-utils-1.0.32-4.fc18 has been submitted as an update for Fedora 18.
https://admin.fedoraproject.org/updates/scsi-target-utils-1.0.32-4.fc18

scsi-target-utils-1.0.32-4.fc18 has been pushed to the Fedora 18 stable repository.  If problems still persist, please make note of it in this bug report.

scsi-target-utils-1.0.24-7.fc17 has been pushed to the Fedora 17 stable repository.  If problems still persist, please make note of it in this bug report.

I have scsi-target-utils-1.0.32-4.fc18.x86_64 installed, but am still seeing this issue.  I don't think the sleep is being executed; the service fails very quickly.

# time systemctl start tgtd.service
Job for tgtd.service failed. See 'systemctl status tgtd.service' and 'journalctl -xn' for details.

real	0m0.138s
user	0m0.003s
sys	0m0.013s

Does it work if the line is changed to "/bin/sleep 5" ?

Andy:

I think it should be '/bin/sleep 5' yes.

Here is a copy of the init script I've been using:

https://github.com/redhat-openstack/openstack-puppet/blob/master/modules/cinder/files/tgtd.service#L12

OK respinning pkgs.

scsi-target-utils-1.0.24-8.fc17 has been submitted as an update for Fedora 17.
https://admin.fedoraproject.org/updates/scsi-target-utils-1.0.24-8.fc17

scsi-target-utils-1.0.32-5.fc18 has been submitted as an update for Fedora 18.
https://admin.fedoraproject.org/updates/scsi-target-utils-1.0.32-5.fc18

ok should be good to go. Turn on updates-testing repo to get these pkgs before they go stable.

Works for me.  Thanks!

Package scsi-target-utils-1.0.24-8.fc17:
* should fix your issue,
* was pushed to the Fedora 17 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing scsi-target-utils-1.0.24-8.fc17'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2013-5441/scsi-target-utils-1.0.24-8.fc17
then log in and leave karma (feedback).

scsi-target-utils-1.0.32-5.fc18 has been pushed to the Fedora 18 stable repository.  If problems still persist, please make note of it in this bug report.

scsi-target-utils-1.0.24-8.fc17 has been pushed to the Fedora 17 stable repository.  If problems still persist, please make note of it in this bug report.

I still see this issue in Fedora 18 though sleep 5 is incorporated in /lib/systemd/system/tgtd.service

<snip>
...
ExecStart=/usr/sbin/tgtd -f $TGTD_OPTS
# see bz 848942. workaround for a race for now.
ExecStartPost=/bin/sleep 5
...
</snip>

scsi-target-utils RPM details
=============================
~# rpm -qa | grep scsi-target-utils
scsi-target-utils-1.0.32-5.fc18.x86_64

Service status
==============
~# systemctl status tgtd.service
tgtd.service - tgtd iSCSI target daemon
          Loaded: loaded (/usr/lib/systemd/system/tgtd.service; disabled)
          Active: failed (Result: core-dump) since Wed, 2013-07-24 18:43:08 IST; 1h 10min ago
         Process: 3585 ExecStop=/usr/sbin/tgtadm --op delete --mode system (code=exited, status=0/SUCCESS)
         Process: 3582 ExecStop=/usr/sbin/tgt-admin --update ALL -c /dev/null (code=exited, status=0/SUCCESS)
         Process: 3581 ExecStop=/usr/sbin/tgtadm --op update --mode sys --name State -v offline (code=exited, status=0/SUCCESS)
         Process: 3565 ExecStartPost=/usr/sbin/tgtadm --op update --mode sys --name State -v offline (code=exited, status=0/SUCCESS)
         Process: 3557 ExecStartPost=/bin/sleep 10 (code=exited, status=0/SUCCESS)
         Process: 3556 ExecStart=/usr/sbin/tgtd -f $TGTD_OPTS (code=dumped, signal=ABRT)
          CGroup: name=systemd:/system/tgtd.service

Jul 24 18:41:27 dhcp-10-10-2-8.helab.bdc tgtd[3556]: librdmacm: couldn't read ABI version.
Jul 24 18:41:27 dhcp-10-10-2-8.helab.bdc tgtd[3556]: librdmacm: assuming: 4
Jul 24 18:41:27 dhcp-10-10-2-8.helab.bdc tgtd[3556]: tgtd: iser_ib_init(3376) Failed to initialize RDMA; load kernel modules?
Jul 24 18:41:27 dhcp-10-10-2-8.helab.bdc tgtd[3556]: tgtd: work_timer_start(146) use timer_fd based scheduler
Jul 24 18:41:27 dhcp-10-10-2-8.helab.bdc tgtd[3556]: tgtd: bs_init(313) use signalfd notification
Jul 24 18:43:08 dhcp-10-10-2-8.helab.bdc tgtd[3556]: tgtd: iscsi/iser.c:3429: iser_ib_release: Assertion `list_empty(&iser_conn_list)' failed.
Jul 24 18:43:08 dhcp-10-10-2-8.helab.bdc tgtd[3556]: CMA: unable to get RDMA device list
Jul 24 18:43:08 dhcp-10-10-2-8.helab.bdc systemd[1]: tgtd.service: main process exited, code=dumped, status=6/ABRT
Jul 24 18:43:08 dhcp-10-10-2-8.helab.bdc systemd[1]: Failed to start tgtd iSCSI target daemon.
Jul 24 18:43:08 dhcp-10-10-2-8.helab.bdc systemd[1]: Unit tgtd.service entered failed state

Fedora 17 changed to end-of-life (EOL) status on 2013-07-30. Fedora 17 is 
no longer maintained, which means that it will not receive any further 
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of 
Fedora please feel free to reopen this bug against that version.

Thank you for reporting this bug and we are sorry it could not be fixed.

Please retry with 1.0.38-1 in updates-testing on Fedora 18, can you still reproduce?

This message is a reminder that Fedora 18 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 18. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as WONTFIX if it remains open with a Fedora 
'version' of '18'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version prior to Fedora 18's end of life.

Thank you for reporting this issue and we are sorry that we may not be 
able to fix it before Fedora 18 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior to Fedora 18's end of life.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Fedora 18 changed to end-of-life (EOL) status on 2014-01-14. Fedora 18 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.

Please mark this issue as closed as it's a wont fix for F17.

Currently running F20, scsi-target-utils-1.0.46-1.fc20.x86_64.

The issue seems to be gone.