+++ This bug was initially created as a clone of Bug #221911 +++ On a xen dom0 (tested just now with 2.6.18-1.2869.fc6xen.x86_64), when snmpd is running, it barfs the following two lines into syslog every 30 seconds. Jan 8 12:47:00 barleycorn snmpd[2230]: netsnmp_assert index == tmp failed if-mib/data_access/interface.c:467 _access_interface_entry_save_name() Jan 8 12:47:00 barleycorn snmpd[2230]: netsnmp_assert __extension__ ({ size_t __s1_len, __s2_len; (__builtin_constant_p (rowreq_ctx->data.ifentry->name) && __builtin_constant_p (ifentry->name) && (__s1_len = strlen (rowreq_ctx->data.ifentry->name), __s2_len = strlen (ifentry->name), (!((size_t)(const void *)((rowreq_ctx->data.ifentry->name) + 1) - (size_t)(const void *)(rowreq_ctx->data.ifentry->name) == 1) || __s1_len >= 4) && (!((size_t)(const void *)((ifentry->name) + 1) - (size_t)(const void *)(ifentry->name) == 1) || __s2_len >= 4)) ? __builtin_strcmp (rowreq_ctx->data.ifentry->name, ifentry->name) : (__builtin_constant_p (rowreq_ctx->data.ifentry->name) && ((size_t)(const void *)((rowreq_ctx->data.ifentry->name) + 1) - (size_t)(const void *)(rowreq_ctx->data.ifentry->name) == 1) && (__s1_len = strlen (rowreq_ctx->data.ifentry->name), __s1_len < 4) ? (__builtin_constant_p (ifentry->name) && ((size_t)(const void *)((ifentry->name) + 1) - (size_t)(const void *)(ifentry->name) == 1) ? __builtin_strcmp (rowreq_c Wadda mess! It totally ruins the S/N ratio in /var/log/messages. And just FYI, here is a diff of the interface names/changes with and without xen kernel. --- ip-addr.2007-01-08 2007-01-08 12:13:33.000000000 -0800 +++ - 2007-01-08 12:44:00.105480000 -0800 @@ -1,10 +1,28 @@ 1: lo: <LOOPBACK,UP,LOWER_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 scope host lo -2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000 - link/ether 00:04:23:d0:53:cc brd ff:ff:ff:ff:ff:ff - inet 10.10.10.202/29 brd 10.10.10.207 scope global eth0 +2: peth0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000 + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff 3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000 link/ether 00:04:23:d0:53:cd brd ff:ff:ff:ff:ff:ff 4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000 link/ether 00:13:72:fe:3c:56 brd ff:ff:ff:ff:ff:ff +5: vif0.0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff +6: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue + link/ether 00:04:23:d0:53:cc brd ff:ff:ff:ff:ff:ff + inet 10.10.10.202/29 brd 10.10.10.207 scope global eth0 +7: vif0.1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff +8: veth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff +9: vif0.2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff +10: veth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff +11: vif0.3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff +12: veth3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop + link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff +13: xenbr0: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue + link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff -- Additional comment from rvokal on 2007-01-12 03:07 EST -- Can you please try to reproduce this issue with the latest rawhide net-snmp? I can't reproduce the above assertion on my Xen test box. -- Additional comment from curtis on 2007-01-12 12:26 EST -- Not reproducible with 5.4-5.fc7.x86_64. This bug appears in fc6 i.e. net-snmp 5.3.1. -- Additional comment from curtis on 2007-01-12 15:06 EST -- Awwww foo! I roll back to fc6 package and now bug is gone. However... # chkconfig snmpd on # shutdown -r now It's baaaack. Even further, upgrade to 5.4 (this time my mirror had the rawhide package you built last night), reboot and the problem persists. So it is reproducible with both 5.3 and 5.4. Just only when started via boot. Is there a race here? Or something screwey with my environment? -- Additional comment from mmarcini on 2007-01-16 09:53 EST -- Altough my messages in /var/log/messages were quite different, the fact is, that the snmpd logs wierd things. This difference may be configuration related, but it is bug anyway. The bug is reporducible. -- Additional comment from rs.spam.futz.org on 2007-02-05 10:13 EST -- Fixes checked in upstream for 5.3.x: -fix (reported elsewhere) xen+x86_64 crash: http://net-snmp.cvs.sourceforge.net/net-snmp/net-snmp/snmplib/snmp_logging.c?r1=5.34.2.5&r2=5.34.2.1&pathrev=V5-3-patches -fix for overly-verbose log message: http://net-snmp.cvs.sourceforge.net/net-snmp/net-snmp/agent/mibgroup/if-mib/ifTable/ifTable_data_access.c?r1=1.18.2.3&r2=1.18.2.4 -- Additional comment from rvokal on 2007-03-12 10:53 EST -- Both pathes are in the latest rawhide version, I believe this issue is fixed in net-snmp-5.4-13
Just wanted to make sure this bug was flagged for RHEL5, as well. From /var/log/messages: Aug 22 15:48:25 lnxvmhost1 snmpd[3857]: netsnmp_assert __extension__ ({ size_t __s1_len, __s2_len; (__builtin_constant_p (rowreq_ctx->data.ifentry->name) && __builtin_constant_p (ifentry->name) && (__s1_len = strlen (rowreq_ctx->data.ifentry->name), __s2_len = strlen (ifentry->name), (!((size_t)(const void *)((rowreq_ctx->data.ifentry->name) + 1) - (size_t)(const void *)(rowreq_ctx->data.ifentry->name) == 1) || __s1_len >= 4) && (!((size_t)(const void *)((ifentry->name) + 1) - (size_t)(const void *)(ifentry->name) == 1) || __s2_len >= 4)) ? __builtin_strcmp (rowreq_ctx->data.ifentry->name, ifentry->name) : (__builtin_constant_p (rowreq_ctx->data.ifentry->name) && ((size_t)(const void *)((rowreq_ctx->data.ifentry->name) + 1) - (size_t)(const void *)(rowreq_ctx->data.ifentry->name) == 1) && (__s1_len = strlen (rowreq_ctx->data.ifentry->name), __s1_len < 4) ? (__builtin_constant_p (ifentry->name) && ((size_t)(const void *)((ifentry->name) + 1) - (size_t)(const void *)(ifentry->name) == 1) ? __builtin_strcmp (rowreq_c ... about twice every 30 seconds.
*** This bug has been marked as a duplicate of 239083 ***
ah, most of the duplicate bug is private and you cannot see it... well, it's just about exactly the same message you are describing. Should be fixed in RHEL 5.1