[Bugs] [Bug 1761365] New: libgfapi: the glfs_init() get stuck and is in inifinitely loop in pthread_spin_lock()
bugzilla at redhat.com
bugzilla at redhat.com
Mon Oct 14 09:19:54 UTC 2019
https://bugzilla.redhat.com/show_bug.cgi?id=1761365
Bug ID: 1761365
Summary: libgfapi: the glfs_init() get stuck and is in
inifinitely loop in pthread_spin_lock()
Product: GlusterFS
Version: 7
Status: NEW
Component: libgfapi
Assignee: bugs at gluster.org
Reporter: xiubli at redhat.com
QA Contact: bugs at gluster.org
CC: bugs at gluster.org
Target Milestone: ---
Classification: Community
Description of problem:
I am now testing the gfapi stuff based on the gluster-block/tcmu-runner, and
hit one problem that the tcmu-runner process is running in almost 100% cpu and
get stuck when creating the gluster-block device:
The gluster-block command is:
[root at localhost tcmu-runner]# gluster-block create repvol/block ha 2 prealloc
full 10.70.39.238,10.70.39.231 1G
[root at localhost tcmu-runner]# top
top - 14:14:50 up 1:07, 2 users, load average: 2.06, 1.89, 1.17
Tasks: 116 total, 2 running, 114 sleeping, 0 stopped, 0 zombie
%Cpu(s): 50.0 us, 3.1 sy, 0.0 ni, 46.9 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 1990.4 total, 853.9 free, 270.9 used, 865.6 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 1560.8 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
8020 root 20 0 2412664 35916 21792 R 93.8 1.8 12:39.00
tcmu-runner
1 root 20 0 108892 15540 9472 S 0.0 0.8 0:02.44 systemd
2 root 20 0 0 0 0 S 0.0 0.0 0:00.01 kthreadd
3 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 rcu_gp
[root at localhost tcmu-runner]# perf top -p 8020
Samples: 7K of event 'cpu-clock:pppH', 4000 Hz, Event count (approx.):
1838750000 lost: 0/0 drop: 0/0
Overhead Shared Object Symbol
99.95% libpthread-2.29.so [.] pthread_spin_lock
0.01% [kernel] [k] __ip_queue_xmit
0.01% [kernel] [k] __softirqentry_text_start
0.01% [kernel] [k] _raw_spin_unlock_irqrestore
0.01% [kernel] [k] run_rebalance_domains
[root at localhost tcmu-runner]# pstack 8020
Thread 17 (Thread 0x7f709a7fc700 (LWP 11351)):
...
Thread 1 (Thread 0x7f7128527880 (LWP 8020)):
#0 0x00007f7128d4f2b5 in pthread_spin_lock () at /lib64/libpthread.so.0
#1 0x00007f7126ba6eba in mem_get () at /lib64/libglusterfs.so.0
#2 0x00007f7126ba6fdd in mem_get0 () at /lib64/libglusterfs.so.0
#3 0x00007f7126b6e004 in get_new_dict_full () at /lib64/libglusterfs.so.0
#4 0x00007f7126b6f9f0 in dict_new () at /lib64/libglusterfs.so.0
#5 0x00007f7126ce9e38 in glfs_init_common () at /lib64/libgfapi.so.0
#6 0x00007f7126cea030 in glfs_init () at /lib64/libgfapi.so.0
#7 0x00007f7126d20332 in tcmu_glfs_unlock () at
/usr/lib64/tcmu-runner/handler_glfs.so
[root at localhost tcmu-runner]#
It is infinitely looping in the libglusterfs.so .......
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. install gluterfs-7.0 packages from
https://download.gluster.org/pub/gluster/glusterfs/qa-releases/7.0rc3/Fedora/fedora-30/x86_64/
and the tcmu-runner/gluster-block from source.
2. enable and start glusterd/tcmu-runner/gluster-blockd services.
3. create one replicate volume:
# gluster vol create repvol replica 2 10.70.39.238:/data/repvol
10.70.39.231:/data/repvol force
# gluster vol set repvol group gluster-block
# gluster vol start repvol
# gluster volume set repvol locks.mandatory-locking forced
#gluster volume set repvol enforce-mandatory-lock on
#gluster volume set repvol performance.client-io-threads off
4. then create the gluster-block device by using:
# gluster-block create repvol/block ha 2 prealloc full
10.70.39.238,10.70.39.231 1G
5. it will be stuck in Step4.
Actual results:
Expected results:
Additional info:
--
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
More information about the Bugs
mailing list