Ubuntu has issued an advisory on May 23: http://www.ubuntu.com/usn/usn-1450-1/ The patch applied upstream to fix this is attached here: https://bugzilla.redhat.com/show_bug.cgi?id=815813 Cauldron/Mageia 2 are also affected.
CC: (none) => guillomovitch
SuSE has issued an advisory for this today (May 29): http://lists.opensuse.org/opensuse-updates/2012-05/msg00042.html
Version: 1 => CauldronWhiteboard: (none) => MGA2TOO, MGA1TOO
RedHat has issued an advisory for this today (June 20): https://rhn.redhat.com/errata/RHSA-2012-0876.html
Mandriva has issued an advisory for this today (June 21): http://www.mandriva.com/en/support/security/advisories/?dis=2010.1&name=MDVSA-2012:099
Patched packages uploaded for Cauldron, Mageia 2, and Mageia 1. Advisory: ======================== Updated net-snmp packages fix security vulnerability: An array index error, leading to out-of heap-based buffer read flaw was found in the way net-snmp agent performed entries lookup in the extension table. When certain MIB subtree was handled by the extend directive, a remote attacker having read privilege to the subtree could use this flaw to cause a denial of service (snmpd crash) via SNMP GET request involving a non-existent extension table entry (CVE-2012-2141). References: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2012-2141 http://www.mandriva.com/en/support/security/advisories/?dis=2010.1&name=MDVSA-2012:099 ======================== Updated packages in core/updates_testing: ======================== net-snmp-5.6.1-7.1.mga1 libnet-snmp25-5.6.1-7.1.mga1 libnet-snmp-devel-5.6.1-7.1.mga1 libnet-snmp-static-devel-5.6.1-7.1.mga1 net-snmp-utils-5.6.1-7.1.mga1 net-snmp-tkmib-5.6.1-7.1.mga1 net-snmp-mibs-5.6.1-7.1.mga1 net-snmp-trapd-5.6.1-7.1.mga1 perl-NetSNMP-5.6.1-7.1.mga1 net-snmp-5.7.1-3.1.mga2 libnet-snmp30-5.7.1-3.1.mga2 libnet-snmp-devel-5.7.1-3.1.mga2 libnet-snmp-static-devel-5.7.1-3.1.mga2 net-snmp-utils-5.7.1-3.1.mga2 net-snmp-tkmib-5.7.1-3.1.mga2 net-snmp-mibs-5.7.1-3.1.mga2 net-snmp-trapd-5.7.1-3.1.mga2 perl-NetSNMP-5.7.1-3.1.mga2 python-netsnmp-5.7.1-3.1.mga2 from SRPMS: net-snmp-5.6.1-7.1.mga1.src.rpm net-snmp-5.7.1-3.1.mga2.src.rpm
Version: Cauldron => 2Assignee: bugsquad => qa-bugsWhiteboard: MGA2TOO, MGA1TOO => MGA1TOO
Testing complete on Mageia 1 i586. No POC, so just testing that smmpd is working ... $ snmpget -c public localhost system.sysDescr.0 SNMPv2-MIB::sysDescr.0 = STRING: "Linux 2.4.x" I'll test Mageia 2 i586 shortly.
CC: (none) => davidwhodgins
Testing complete on Mageia 2 i586.
Whiteboard: MGA1TOO => MGA1TOO, mga1-32-OK mga2-32-OK
Testing complete x86_64 mga2 # service snmpd start Starting snmpd (via systemctl): [ OK ] # snmpget -c public localhost system.sysDescr.0 SNMPv2-MIB::sysDescr.0 = STRING: "Linux 2.4.x" Should this be reporting Linux 2.4.x ?
Hardware: i586 => AllWhiteboard: MGA1TOO, mga1-32-OK mga2-32-OK => MGA1TOO, mga1-32-OK mga2-32-OK mga2-64-OK
Testing complete x86_64 mga1 If 2.4.x is correct then this can be validated.
Whiteboard: MGA1TOO, mga1-32-OK mga2-32-OK mga2-64-OK => MGA1TOO, mga1-32-OK mga2-32-OK mga2-64-OK mga1-64-OK
My goodness! This comes from the snmpd.conf file that the package installs. This file hasn't been modified at all since before Mandriva's SVN was created in 2007, and probably not since it was first copied from RedHat/Fedora many many years ago. That string is really just put into the config file as a generic placeholder, and the sysadmin is expected to change it to whatever they want it to say for that machine. Obviously it's not "correct" so to speak, but it's not anything we need to hold up this update for. At some point, someone should go through and re-sync a lot of the sources in this package with Fedora circa 2012 :o), and Mandriva needs to do this as well, since theirs is the same as ours. I guess this is what happens when a package isn't well maintained for years, or when separate sources from upstream are copied from somewhere else. We should do that re-sync in Cauldron, but there's probably more risk than benefit in trying to do it for these updates, so I think they're fine as-is for now. Thanks for catching this.
Thanks for checking, I thought it was odd :) Validating then Please see comment 4 for advisory and srpms Updates for mga1 and mga2 Could sysadmin please push from core/updates_testing to core/updates Thanks!
Keywords: (none) => validated_updateCC: (none) => sysadmin-bugs
Update pushed: https://wiki.mageia.org/en/Support/Advisories/MGASA-2012-0128
Status: NEW => RESOLVEDCC: (none) => tmbResolution: (none) => FIXED