Note: This bug is displayed in read-only format because
the product is no longer active in Red Hat Bugzilla.
RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Reproducer:
1. Run the script
---8<---8<---
#!/bin/ksh
function funcTERM
{
exit 0
}
trap funcTERM SIGTERM
while true
do
i=0
while [ $i -lt 100 ]
do
DUMMY=`echo dummy`
let ++i
done
ps u --no-headers -p $$
sleep 0.1
done
---8<---8<---
2. Keep looking at 6th column(RSS) of ps
# ./check.ksh
root 8643 0.0 0.3 109344 1668 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh
...
root 8643 4.2 0.4 109600 2136 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh
root 8643 4.3 0.4 109600 2148 pts/1 R+ 13:19 0:00 /bin/ksh ./check.ksh
root 8643 4.3 0.4 109600 2148 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh
Minimal patch to correct the problem:
"""
diff -up ksh-20120801/src/cmd/ksh93/sh/subshell.c.orig ksh-20120801/src/cmd/ksh93/sh/subshell.c
--- ksh-20120801/src/cmd/ksh93/sh/subshell.c.orig 2018-05-11 13:45:12.280999698 -0300
+++ ksh-20120801/src/cmd/ksh93/sh/subshell.c 2018-05-11 13:46:19.396000044 -0300
@@ -575,7 +575,7 @@ Sfio_t *sh_subshell(Shell_t *shp,Shnode_
sp->coutpipe = shp->coutpipe;
sp->cpipe = shp->cpipe[1];
shp->cpid = 0;
- sh_sigreset(0);
+ sh_sigreset(1);
}
jmpval = sigsetjmp(buff.buff,0);
if(jmpval==0)
"""
The patch could be made a bit longer, but more complex, to avoid the strdup
calls that will be just released shortly after (with this patch).
** Note ** the equivalent patch is required in xec.c, but there is no reproducer
so far for the leak, that would happen in the same conditions (sh_sigreset zero'ing
the vector of strdup'ed strings).
Created attachment 1436213[details]
ksh-20120801-subshell-leak.patch
Indeed it is a dup.
But please note that the patch in #1460940 will again crash with the reproducer in
#1117404, and almost certainly in another condition with functions (that I do not have
ready the reproducer right now...).
This new patch is a revised version of the (likely) "proper" patch, that depends on
what sh_sigreset does. The previous patch, that was causing the leak, was incorrectly
addressing the real problem (keeping pointers to released strings).
The root cause of the problem is sh_sigreset free'ing strings, and the patch in
#1460940 (a revert) just causes it to again keep using released strings. The crash
condition in the other reports are rooted on double frees (as far as I understand it).
Comment 4Siteshwar Vashisht
2018-05-15 13:11:47 UTC
> But please note that the patch in #1460940 will again crash with the reproducer in #1117404, and almost certainly in another condition with functions (that I do not have ready the reproducer right now...).
You are right. I am going to add your patch to bug 1460940.
Reproducer: 1. Run the script ---8<---8<--- #!/bin/ksh function funcTERM { exit 0 } trap funcTERM SIGTERM while true do i=0 while [ $i -lt 100 ] do DUMMY=`echo dummy` let ++i done ps u --no-headers -p $$ sleep 0.1 done ---8<---8<--- 2. Keep looking at 6th column(RSS) of ps # ./check.ksh root 8643 0.0 0.3 109344 1668 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh ... root 8643 4.2 0.4 109600 2136 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh root 8643 4.3 0.4 109600 2148 pts/1 R+ 13:19 0:00 /bin/ksh ./check.ksh root 8643 4.3 0.4 109600 2148 pts/1 S+ 13:19 0:00 /bin/ksh ./check.ksh Minimal patch to correct the problem: """ diff -up ksh-20120801/src/cmd/ksh93/sh/subshell.c.orig ksh-20120801/src/cmd/ksh93/sh/subshell.c --- ksh-20120801/src/cmd/ksh93/sh/subshell.c.orig 2018-05-11 13:45:12.280999698 -0300 +++ ksh-20120801/src/cmd/ksh93/sh/subshell.c 2018-05-11 13:46:19.396000044 -0300 @@ -575,7 +575,7 @@ Sfio_t *sh_subshell(Shell_t *shp,Shnode_ sp->coutpipe = shp->coutpipe; sp->cpipe = shp->cpipe[1]; shp->cpid = 0; - sh_sigreset(0); + sh_sigreset(1); } jmpval = sigsetjmp(buff.buff,0); if(jmpval==0) """ The patch could be made a bit longer, but more complex, to avoid the strdup calls that will be just released shortly after (with this patch). ** Note ** the equivalent patch is required in xec.c, but there is no reproducer so far for the leak, that would happen in the same conditions (sh_sigreset zero'ing the vector of strdup'ed strings).