System hung due to blocked tasks of IBM Infosphere Datastage Engine
Issue
- jbd2/dm-11-8 high utilization, cannot run sosreport
-
osh processes of IBM DataStage Engine parallel jobs going into "D" state
-
When the VM guest is vmotioned to a different hypervisor the issue goes away for a few hours and occurs again when the osh tasks (of IBM Infosphere DataStage Engine) gets into un-interruptable D state and the system was getting hung and unresponsive.
crash> ps -m | grep UN
[0 00:00:00.012] [UN] PID: 23191 TASK: ffff88057b2bb520 CPU: 2 COMMAND: "chgrp"
[0 00:00:00.013] [UN] PID: 2106 TASK: ffff8809289a2040 CPU: 1 COMMAND: "jbd2/dm-12-8"
[0 00:00:00.013] [UN] PID: 30054 TASK: ffff88092bf7cab0 CPU: 1 COMMAND: "flush-253:12"
[0 00:00:04.164] [UN] PID: 17412 TASK: ffff88089e7c4040 CPU: 1 COMMAND: "osh"
[0 00:00:04.696] [UN] PID: 20503 TASK: ffff8808771a0040 CPU: 1 COMMAND: "osh"
[0 00:00:04.743] [UN] PID: 20630 TASK: ffff880825b68ab0 CPU: 1 COMMAND: "osh"
[0 00:00:22.565] [UN] PID: 27524 TASK: ffff88092cc5b520 CPU: 3 COMMAND: "osh"
[0 00:00:26.847] [UN] PID: 27443 TASK: ffff88091e97a040 CPU: 2 COMMAND: "osh"
[0 00:00:27.753] [UN] PID: 20031 TASK: ffff8804c194c040 CPU: 2 COMMAND: "osh"
[0 00:00:34.031] [UN] PID: 20421 TASK: ffff8805f6bb9520 CPU: 0 COMMAND: "osh"
[0 00:00:35.467] [UN] PID: 18698 TASK: ffff88011e6de040 CPU: 3 COMMAND: "osh"
[0 00:00:39.865] [UN] PID: 19404 TASK: ffff88071947b520 CPU: 0 COMMAND: "osh"
[0 00:00:46.310] [UN] PID: 20336 TASK: ffff880731ae3520 CPU: 1 COMMAND: "osh"
[0 00:00:50.430] [UN] PID: 17656 TASK: ffff88011e6df520 CPU: 1 COMMAND: "osh"
[0 00:00:54.730] [UN] PID: 13791 TASK: ffff880737d5b520 CPU: 1 COMMAND: "osh"
[0 00:01:09.968] [UN] PID: 1137 TASK: ffff8808872e1520 CPU: 1 COMMAND: "osh"
[0 00:01:18.562] [UN] PID: 26692 TASK: ffff880102becab0 CPU: 3 COMMAND: "osh"
[0 00:01:30.395] [UN] PID: 22222 TASK: ffff880923f64040 CPU: 1 COMMAND: "osh"
[0 00:01:33.840] [UN] PID: 12621 TASK: ffff880146eeeab0 CPU: 2 COMMAND: "osh"
[0 00:01:36.298] [UN] PID: 11269 TASK: ffff8805f6a2a040 CPU: 1 COMMAND: "osh"
[0 00:01:41.918] [UN] PID: 20785 TASK: ffff88091cb1eab0 CPU: 0 COMMAND: "osh"
[0 00:01:53.106] [UN] PID: 24394 TASK: ffff8808a4696ab0 CPU: 2 COMMAND: "osh"
[0 00:01:58.707] [UN] PID: 18383 TASK: ffff88092e9c0040 CPU: 2 COMMAND: "osh"
crash>
Environment
- Red Hat Enterprise Linux 6.7
- Vmware
IBM Infosphere Datastage Software(high performance parallel ETL jobs processing Engine)
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.