Red Hat Bugzilla – Full Text Bug Listing
|Summary:||SIGSEGV in krb5kdc at foreachaddr.c:205 (NULL reference)|
|Product:||[Fedora] Fedora||Reporter:||Kuba Ober <kuba>|
|Component:||krb5||Assignee:||Nalin Dahyabhai <nalin>|
|Status:||CLOSED UPSTREAM||QA Contact:|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2006-07-12 09:16:21 EDT||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Kuba Ober 2004-01-12 17:25:41 EST
Description of problem: There's a bug in src/include/foreachaddr.c. It's manifests itself around line 205, although the bug is really present in a few places. The ifp->ifaddr may be zero, and they get referenced in ifequal -> SIGSEGV. Version-Release number of selected component (if applicable): 1.3.1-7 How reproducible: Happens upon krb5kdc startup (only), with following network interfaces: eth0 lo ppp0 (ifconfig lists them in that order) Steps to Reproduce: 1. service krb5kdc start Actual results: FAILED Expected results: OK Additional info: A clean /var/log/krb5kdc.log shows (after krb5kdc bombed out) setting up network... setting up network... skipping unrecognized local address family 17 skipping unrecognized local address family 17
Comment 1 Kuba Ober 2004-01-12 17:27:27 EST
Created attachment 96908 [details] A makeshift fix which makes the problem go away The attached patch fixes the problem for me. It's a makeshift fix, the problem should be investigated and a proper solution found.
Comment 2 Kuba Ober 2004-01-12 17:31:39 EST
It may be useful to know when did the error manifest itself: A working, up-to-date RedHat 9 server with running krb5kdc was upgraded to Fedora Core 1. After updating all the packages, as well as applying the updated modules, the system was rebooted. krb5kdc didn't want to start up (showed [FAILED]). FYI: This is a stable system, rpm -Va passes w/o anything out of the ordinary, memtest86 w/extensive tests OK for 96 hours. Nothing unusual I guess. IPV4 only, redhat .i686 kernel, everything really stock.
Comment 3 Kuba Ober 2004-01-12 17:33:08 EST
Obviously, the krb5 rpm version is 1.3.1-6 :)
Comment 4 Nalin Dahyabhai 2004-01-19 14:17:50 EST
*** Bug 113558 has been marked as a duplicate of this bug. ***
Comment 5 Nalin Dahyabhai 2004-01-19 15:44:37 EST
I've only been able to reproduce this in a weird setup (with a network connection up, but configured without an address). Can you please append the output of either 'ip address list' or 'ifconfig -a' so that I can make sure that's what's going on for you as well?
Comment 6 Kuba Ober 2004-01-19 16:18:03 EST
This is my network setup where the problem reproduceably recurs. Note that this system runs fedora kernel and thus has IPV6 available, although not necessarily configured. I didn't do anything even remotely IPV6 related on this machine, though. 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo 3: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:d0:b7:17:39:3e brd ff:ff:ff:ff:ff:ff inet 192.168.0.1/24 brd 192.168.0.255 scope global eth0 4: ppp0: <POINTOPOINT,MULTICAST,NOARP,UP> mtu 1656 qdisc pfifo_fast qlen 3 link/ppp inet 126.96.36.199/32 scope global ppp0 eth0 Link encap:Ethernet HWaddr 00:D0:B7:17:39:3E inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:13891679 errors:0 dropped:0 overruns:0 frame:0 TX packets:14152475 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:591735880 (564.3 Mb) TX bytes:1513154826 (1443.0 Mb) Interrupt:12 Base address:0xcc00 Memory:dffdf000-dffdf038 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:290810 errors:0 dropped:0 overruns:0 frame:0 TX packets:290810 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:92174788 (87.9 Mb) TX bytes:92174788 (87.9 Mb) ppp0 Link encap:Point-to-Point Protocol inet addr:188.8.131.52 P-t-P:184.108.40.206 Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1656 Metric:1 RX packets:1133681 errors:53 dropped:0 overruns:0 frame:0 TX packets:933852 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:3 RX bytes:563438740 (537.3 Mb) TX bytes:110723212 (105.5 Mb)
Comment 7 Nalin Dahyabhai 2004-02-03 16:33:15 EST
Hmm. By the looks of it, you shouldn't even be hitting that problem on your system -- every interface has at least one network address. Please try 1.3.1-9 from the development tree, which has a patch which is very similar to the one submitted in http://mailman.mit.edu/pipermail/krb5-bugs/2004-January/002152.html
Comment 8 Paul Jakma 2004-02-03 20:17:56 EST
His interfaces all have v4 addresses, however they might not have v6 addresses, which appears to be the source of the problem - eg the sit0 device is a common case. The patch at the URL above fixes the problem here.
Comment 9 Matthew Miller 2006-07-11 13:46:29 EDT
Fedora Core 1 is maintained by the Fedora Legacy project for security updates only. If this problem is a security issue, please reopen and reassign to the Fedora Legacy product. If it is not a security issue and hasn't been resolved in the current FC5 updates or in the FC6 test release, reopen and change the version to match. Thanks! NOTE: Fedora Core 1 is reaching the final end of support even by the Legacy project. After Fedora Core 6 Test 2 is released (currently scheduled for July 26th), there will be no more security updates for FC1. Please use these next two weeks to upgrade any remaining FC1 systems to a current release.
Comment 10 Paul Jakma 2006-07-12 05:44:14 EDT
Long fixed in MIT Krb5 sources.
Comment 11 Matthew Miller 2006-07-12 09:16:21 EDT
So, resolved upstream?