[Ocfs2-devel] one issue, someone may give some help,thanks

[Guozhonghua]

　　Hi， I am using ocfs2 for test and find one issue, would someone give some help about it.

　　

　　The background is that we are using ocfs2 over iSCSI on Ubuntu 12.04, with kernel version 3.2.0-23.

　　Linux 3.2.0-23-generic x86_64 GNU/Linux

　　

　　We changed the configuration file of iSCSI, initiatorname.iscsi file, so as to restart the iSCSI service with the command "service open-iscsi restart".

　　After the restart of iscsi service, the device that ocfs2 using is missed, and the log is written to kern.log as below:

　　Jun  5 09:10:06 vmc-7 kernel: [45946.187897] sd 6:0:0:13: [sde] Attached SCSI disk

　　Jun  5 09:10:07 vmc-7 kernel: [45946.439225] (o2hb-692FC96E2F,2786,5):o2hb_bio_end_io:389 ERROR: IO Error -19

　　Jun  5 09:10:07 vmc-7 kernel: [45946.439673] (o2hb-692FC96E2F,2786,5):o2hb_do_disk_heartbeat:1017 ERROR: status = -19

　　Jun  5 09:10:07 vmc-7 kernel: [45947.117979] (o2hb-58592532BC,2804,2):o2hb_bio_end_io:389 ERROR: IO Error -19

　　Jun  5 09:10:07 vmc-7 kernel: [45947.118413]

　　... ... .... ...

　　 (o2hb-58592532BC,2804,0):o2hb_bio_end_io:389 ERROR: IO Error -19

　　Jun  5 09:10:53 vmc-7 kernel: [45993.143578] (o2hb-58592532BC,2804,0):o2hb_do_disk_heartbeat:1017 ERROR: status = -19

　　

　　The heartbeat timeout is 31 seconds configured as below.

　　# O2CB_HEARTBEAT_THRESHOLD: Iterations before a node is considered dead.

　　O2CB_HEARTBEAT_THRESHOLD=31

　　

　　root at vmc-7:/sys/kernel/config/cluster/iPool/heartbeat# cat dead_threshold

　　31

　　

　　Why elapse more than 50 seconds, the server does not reset?

　　

　　I review the code of kernel with version 3.2.2, and puzzled with heartbeat code as below, is it right that the loop does not break out so as the server does not reset.

　　static int o2hb_thread(void *data)

　　{

　　 ... ... ...

　　

　　 while (!kthread_should_stop() &&

　　        !reg->hr_unclean_stop && !reg->hr_aborted_start) {

　　

　　  }

　　}

　　

　　In the function o2hb_do_disk_heartbeat(struct o2hb_region *reg), while

　　ret = o2hb_read_slots(reg, highest_node + 1); error with 19, ENODEV, the device may be does not existed for the iSCSI service restart.

　　

　　Is anything wrong with it?

　　

　　Thanks.

　　

-------------------------------------------------------------------------------------------------------------------------------------
本邮件及其附件含有杭州华三通信技术有限公司的保密信息，仅限于发送给上面地址中列出
的个人或群组。禁止任何其他人以任何形式使用（包括但不限于全部或部分地泄露、复制、
或散发）本邮件中的信息。如果您错收了本邮件，请您立即电话或邮件通知发件人并删除本
邮件！
This e-mail and its attachments contain confidential information from H3C, which is
intended only for the person or entity whose address is listed above. Any use of the
information contained herein in any way (including, but not limited to, total or partial
disclosure, reproduction, or dissemination) by persons other than the intended
recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender
by phone or email immediately and delete it!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://oss.oracle.com/pipermail/ocfs2-devel/attachments/20120605/90b93aab/attachment-0001.html