We needed to reset the NSPR error code before calling PR_Read().
Verification procedure: 1. Install a TPS 2. On the TPS UI, on the agent tab disable the userkey profile. This profile is large - so saving it will create large posts. 3. In the admin tab, edit this profile and save your changes, confirm the changes etc. 4. Repeat step 3 multiple times. If the bug is still there, then at some point, the cpu of httpd will go very high (lie 100%) and will stay there for a long time. ( This used to happen after the profile was saved maybe a couple of times.)
Verified on RHEL5.6(x86_64) From comment#4, after disabling 'userKey' profile I repeatedly edited several times & saved it. The %CPU utilization never exceeded 25%.(like below) ################################################################ [root@cs81autoel5 ~]# top -p29971 -p29990 -p15607 PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 29971 root 18 0 295m 58m 3592 S 24.3 5.4 0:00.71 httpd.worker 29990 pkiuser 25 0 994m 61m 4072 S 0.0 5.6 0:04.65 httpd.worker 15607 apache 25 0 199m 3576 728 S 0.0 0.3 0:00.00 httpd [root@cs81autoel5 ~]# ################################################################ Version: mod_nss-1.0.8-2.el5 Build : CS nightly build(Oct3)
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2011-0049.html