Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1631786

Summary: mount hung, all threads blocked
Product: [Retired] nfs-ganesha Reporter: Kaleb KEITHLEY <kkeithle>
Component: MainNFSDAssignee: Frank Filz <ffilz>
Status: CLOSED WONTFIX QA Contact:
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 2.7CC: dang, pasik
Target Milestone: ---   
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: 2020-06-23 12:51:08 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:
Attachments:
Description Flags
backtrace of all threads. none

Description Kaleb KEITHLEY 2018-09-21 14:50:52 UTC
Created attachment 1485628 [details]
backtrace of all threads.

Description of problem:

ganesha/gdb crashed when exiting so we were unable to debug further

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

nfs-ganesha-2.7.0


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 1 Daniel Gryniewicz 2018-09-21 15:06:50 UTC
So, here's what I was able to work out from the backtrace:

Most real threads are blocked on the host mutex in nsm_monitor().  One thread (LWP 13570) has that lock, and is blocked on the rec dplx lock in clnt_req_setup().  I wasn't able to find any thread that has that lock, nor any codepaths that might return without releasing that lock (even in error paths).  Unfortunately, we weren't able to look in the debugger to see which thread has it locked, so we're kind of stuck.

This is clearly an NFSv3 lock-heavy workload.  Beyond that, there's not much I can say.

One possibility is that the rec was destroyed, and the lock is in a bad state, but that's likely to cause a crash, rather than a hang.

Comment 2 Kaleb KEITHLEY 2020-06-23 12:51:08 UTC
open a new bug if seen in 2.8.x or 3.x