Location: on the road for work; home is private time
Posts: 456
Thanks Given: 10
Thanked 108 Times in 100 Posts
the fullcore setting - which is what the syslog/errpt message was about are more specific to application/system 'crashes' that result in a machine-binary core dump.
Java has it's own style of coredumps, text if I recall, and so I suspect they fall under the fsize setting, rather than coresize.
Under normal circumstances you should not be having a lot of coredumps - so setting it unlimited should not be a concern. If you are debugging an issue having an incomplete dump (that is maybe hard to generate) is a tremendous headache - again I would go for unlimited.
When you doubt you will ever use a core for analysis - then you could keep them sized smaller than unlimited.
My opinion, hope it helps.
This User Gave Thanks to MichaelFelt For This Post:
there are two types of cores in conjunction with Java:
- "normal" AIX cores
- Java cores and heapdumps
fullcore and ulimit regulates "normal" AIX cores, not Java cores. If a java process dumps, it can produce just java dump and no AIX core file. In this case it doesn't matter if fullcore=yes and ulimit -с = unlimited. But it can also be the other case - the process (because it is AIX process) dumps a core and Java has no idea about this dump and doesn't write its own dump. In this case it is better to have fullcore=yes and ulimit -c = unlimited.
But before setting ulimit (or better - chuser core=-1 username ;-) ask your AIX administrator. In many environments it is forbidden to produce full core dumps because of security and because the filesystem can be overflowed with dumps, if you forget to remove them.
When I tried to configure GNU make, I received:...
WARNING: Your system has neither waitpid() nor wait3().
Without one of these, signal handling is unreliable
You should be aware that running GNU make with -j
could result in erratic behavior.
...
What is that supposed to mean ? my spec:
... (1 Reply)
I'm running CentOS 6.8 and use bash. I would like a warning to appear to the user who runs the command "service httpd restart"
E.g.
# service httpd restart
are you sure y/n
n
#
(or if y, the command executes).
I looked into it a little but am not sure of the best approach. Aliases I... (2 Replies)
I'm running CentOS 6.8 and use bash. I would like a warning to appear to the user who runs the command "service httpd restart"
E.g.
# service httpd restart
are you sure y/n
n
#
(or if y, the command executes).
I looked into it a little but am not sure of the best approach. Aliases I ... (1 Reply)
(Apologies for any typos.)
OSX 10.12.3 AND Windows 10.
This is for the serious Python experts on at least 3.5.x and above...
In script format sys.stdout.write() AND sys.stderr.write() seems to work correctly.
Have I found a serious bug in the interactive sys.stdout.write() AND... (2 Replies)
Hi all
I have a newly installed Oracle X2-4 server running Solaris 10 x86 with the latest patches.
I have one non-global zone configured running an Oracle DB instance.
After configuring IPMP failover between two NICs on the server and rebooting I am seeing the /var/adm/messages being flooded... (7 Replies)
hi guys
I have suse 11 sp1 and I have a lot of warn file filling / these are under /var/log
there's this big one
-rw-r----- 1 root root 3.9G Feb 1 10:28 warn
warn: ASCII text
and the others that are about 2.5 to 3MB - they are about 130 warn-*.bz2
-rw-r----- 1 root root 3.9G Feb... (2 Replies)
I am not able to find warn-codes that should be used in
#pragma warn -<code>
directive!:wall:
Could anybody advise where I can see a list of warnings with codes that (as I understand) should be 3-letters code?
I have a pro-C program that produces some warnings.
(Do not advise,... (4 Replies)
(I will not duplicate my post that I create in 'Programming' ( My post ), but the issue also (after C ) is related to Sun Solaris.)
I need to find the warning-codes to be used in the
#pragma warn..
C-code directives to suppress some compilation warnings.
More desciptive explanation you... (2 Replies)
As a UNIX newbie, how can I create a (cron)script that rotates my syslogs on AIX 4.3.3 on a 24 hour basis and compresses the old logs ?
TIA ! (1 Reply)