Bug 8859

Summary: postgresql cause random hard system crashes on Ultra5 and Ultra10
Product: [Retired] Red Hat Linux Reporter: Steevithak <steevithak>
Component: kernelAssignee: Michael K. Johnson <johnsonm>
Status: CLOSED WORKSFORME QA Contact:
Severity: high Docs Contact:
Priority: medium    
Version: 6.1   
Target Milestone: ---   
Target Release: ---   
Hardware: sparc   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2000-07-22 19:27:02 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Steevithak 2000-01-25 17:36:18 UTC
Using postgreSQL on RH61 on a Sun Ultra5 or Ultra10 (possibly other sparcs)
results in random hard crashes. These are major crashes - stop-A and the
reset switch don't work, the system must be powered down and restarted and
data loss on the hard disk can occur. A search on deja news reveals that
others are having the same results using other Linux distributions. It
seems to be a kernel related problem rather than a problem in postgreSQL
but I don't have the ability to debug it. At least one user who was having
crashes on a sun Ultra5 running Debian found that upgrading to the newest
kernel solved the problem. Running the regression test will usually cause
the crash, otherwise during normal use, the database causes the crash at
unpredictable intervals - in one case it ran for several days in another it
crashed hard two time in the same hour.

Comment 1 Trond Eivind Glomsrxd 2000-04-06 22:54:59 UTC
Nothing should crash the kernel, so I've reorganized it. Does this happen with
the 6.2 kernel, BTW?

Comment 2 Alan Cox 2000-08-08 19:11:50 UTC
Closed due to no reply


Comment 3 openshift-github-bot 2016-06-22 17:03:00 UTC
Commit pushed to master at https://github.com/openshift/origin

https://github.com/openshift/origin/commit/461ce39506a4c99d250affc5b120ee8e2e1c56b7
Unmarshal ports given in integer format

When docker-compose had a port number written in integer
format, while conversion, it was not being marshalled to
string format, added a type check which will do the int to
string conversion.

Fixes issue #8859

Comment 4 Steevithak 2016-06-22 18:01:39 UTC
Wow! Wasn't expecting to see an update email on a bug I filed 16 years ago on RedHat 6.1! lol Fwiw, I haven't run RH on a sparc for a decade or so. :)