Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
In a HP-UX Serviceguard cluster using the Enterprise Cluster Master Toolkit (ECMT) to configure a highly available CIFS package, if CIFS version A.03.02.04 is installed, upon a failure of node1 currently running the package, the package may not start on node2. Errors similar to the following will be displayed in the syslog log: Aug 20 14:03:15 node2 cmcld[4412]: Timed out unhealthy member(s). Aug 20 14:03:25 node2 cmcld[4412]: Request from node node2 to start package ha_samba on node node2. Aug 20 14:03:25 node2 cmcld[4412]: Executing '/etc/cmcluster/scripts/mscripts/master_control_script.sh start' for package ha_samba, as service PKG*22018. Aug 20 14:03:25 node2 cmserviced[4430]: Request to start package ha_samba Aug 20 14:03:26 node2 smbd[10034]: ERROR: can't open /samba/ha_samba/lock/smbd-smb.conf.pid: Error was File exists Aug 20 14:03:29 node2 cmserviced[4430]: Package Script for ha_samba failed with an exit(1). Aug 20 14:03:29 node2 cmcld[4412]: Package ha_samba run script exited with NO_RESTART. The package log will show an error similar to the following: Aug 20 14:03:25 root@node2 master_control_script.sh[9385]: ###### Starting package ha_samba ###### Aug 20 14:03:25 root@node2 package_ip.sh[9445]: Adding IP address 10.1.1.1 to subnet 10.1.0.0 Aug 20 14:03:25 root@node2 tkit_module.sh[9490]: Install directory mode of operation selected. Aug 20 14:03:25 - Node "node2": Starting Samba daemons Aug 20 14:03:29 - Node "node2": ERROR: Function start_samba_server: Failed to start SAMBA daemons. This occurs because of an anomaly in the way the smbd daemon handles an existing .pid file. If the package is gracefully halted on node1, the file is removed, but if the node fails unexpectedly due to a power loss or unexpected hardware failure, the .pid file will remain on shared storage when the package starts on the adoptive node2. Prior to CIFS A.03.02.04, the sbmd daemon would remove the .pid file if it exists, but in A.03.02.04, an error occurs instead and it fails to run, leading to package start failure. The Samba bug is https://bugzilla.samba.org/show_bug.cgi?id=9505
Any HPE Integrity server as listed in the Products section below and running: Serviceguard A.11.20 for HP-UX Serviceguard Enterprise Cluster Master Toolkit B.07.00.01 CIFS A.03.02.04
Update to CIFS A.04.05.03 or later to get the fix for smbd to restore the previous behavior, where it will remove the .pid file and allow package startup to continue. CIFS A.04 can be obtained from the current HP-UX 11i v2 2017 Update Release media or online from the HP-UX Software Depot home: https://h20392.www2.hpe.com/portal/swdepot/index.do Prior to updating to CIFS A.04.05.03 or later, it may be necessary to update your package configuration. Refer to Customer Advisory a0003745en_us, and be prepared to make the changes documented in that advisory if necessary, when updating to CIVS A.04.05.03 (or later): https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00037456en_us Contact HPE Support if there are any specific questions, or if the workaround described in the advisory above is needed after the CIFS software is updated. Click on the following URL to locate the HPE Customer Support phone number in your country: https://h20195.www2.hpe.com/v2/Getdocument.aspx?docname=A00039121ENW . RECEIVE PROACTIVE UPDATES : Receive support alerts (such as Customer Advisories), as well as updates on drivers, software, firmware, and customer replaceable components, proactively via e-mail through HPE Subscriber's Choice. Sign up for Subscriber's Choice at the following URL: Proactive Updates Subscription Form. NAVIGATION TIP : For hints on navigating HPE.com to locate the latest drivers, patches, and other support software downloads for ProLiant servers and Options, refer to the Navigation Tips document . SEARCH TIP : For hints on locating similar documents on HPE.com, refer to the Search Tips Document .