Bug 1577791

Summary:	tcmu-runner: allow userspace to reset netlink
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	Xiubo Li <xiubli>
Component:	tcmu-runner	Assignee:	Xiubo Li <xiubli>
Status:	CLOSED DUPLICATE	QA Contact:	Rahul Hinduja <rhinduja>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	cns-3.10	CC:	alexander, bgoyal, pkarampu, prasanna.kalever, rhs-bugs, sankarshan, vbellur, xiubli
Target Milestone:	---	Keywords:	Bugfix
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2018-09-10 08:12:51 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1560418
Bug Blocks:

Description Xiubo Li 2018-05-14 06:58:03 UTC

Description of problem:

When the tcmu-runner in userspace is dead or not running, the targetcli process will stuck for every with the following Call Trace: 

2528 [  961.477971] INFO: task targetctl:22655 blocked for more than 120 seconds.
2529 [  961.479456] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
2530 [  961.480171] targetctl       D ffff880169718fd0     0 22655  17249 0x00000080
2531 [  961.480175] Call Trace:
2532 [  961.480186]  [<ffffffff816ab6d9>] schedule+0x29/0x70
2533 [  961.480189]  [<ffffffff816a90e9>] schedule_timeout+0x239/0x2c0
2534 [  961.480195]  [<ffffffff81574d42>] ? skb_release_data+0xf2/0x140
2535 [  961.480197]  [<ffffffff816aba8d>] wait_for_completion+0xfd/0x140
2536 [  961.480203]  [<ffffffff810c6440>] ? wake_up_state+0x20/0x20
2537 [  961.480227]  [<ffffffffc0159f5a>] tcmu_netlink_event+0x26a/0x3a0 [target_core_user]
2538 [  961.480234]  [<ffffffff810b34b0>] ? wake_up_atomic_t+0x30/0x30
2539 [  961.480238]  [<ffffffffc015a2c6>] tcmu_configure_device+0x236/0x350 [target_core_user]
2540 [  961.480257]  [<ffffffffc05085df>] target_configure_device+0x3f/0x3b0 [target_core_mod]
2541 [  961.480264]  [<ffffffffc0502e7c>] target_core_store_dev_enable+0x2c/0x60 [target_core_mod]
2542 [  961.480269]  [<ffffffffc0501244>] target_core_dev_store+0x24/0x40 [target_core_mod]
2543 [  961.480276]  [<ffffffff8128a0e4>] configfs_write_file+0xc4/0x130
2544 [  961.480281]  [<ffffffff81202aed>] vfs_write+0xbd/0x1e0
2545 [  961.480283]  [<ffffffff812038ff>] SyS_write+0x7f/0xe0
2546 [  961.480290]  [<ffffffff816b89fd>] system_call_fastpath+0x16/0x1b

This is because the TCMU in kernel will send some requests to tcmu-runner daemon via netlink by setting the targetcli process into UNINTERRUTABLE state and then goto to sleep. Since the tcmu-runner is not running so will no request be processed and replied.


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 5 Prasanna Kumar Kalever 2018-09-10 08:12:51 UTC


*** This bug has been marked as a duplicate of bug 1476730 ***