birdseye51
Technical User
Please read details below. This has caused V6 systems to reload.
“Leap second handling” on SUSE/NOVELL Enterprise Linux (SLES11) based Large Converged products
Message Body:
SEN LC Products based on SUSE/NOVELL Enterprise Linux (SLES11) can be affected by “leap second handling”.
There is a chance that the kernel leap second notice may cause an overload with a blockage (high CPU load/utilization) and a wide range of effects could be noticed, this could be everything from licensing issues (CLA) to an unstable system, e.g. RMX restarts/reloads or LINUX reboot.
How to verify if your system is affected?
To verify if your system is affected please check the /var/log/messages regarding the following entry:
Jul 1 01:59:59 linux kernel: [103264.364034] Clock: inserting leap second 23:59:60 UTC
If this message is present and your system/kernel has not been rebooted since this event we recommend to plan a service window to perform a system reboot via LINUX command: “reboot”.
Important Note: LINUX systems should be restarted via controlled reboot therefore POWER OFF must be avoided as usual.
Attention: In case your system was affected and it performed a complete LINUX reboot after 01.07.2012/23:59 (UTC) then no further action is needed because the reboot itself resolved the instability issues.
This behavior reflects the current status of analysis of GVS/ Development which is still ongoing.
Which products may be affected?
Following products are using affected LINUX kernel versions and could be impacted when a NTP Service is in operation:
HiPath 4000 V6 systems
Softgate V5
Softgate V6
How to perform recommended system reboot with HiPath4000 V6 Duplex deployment?
a.) Duplex system
1. Reboot Secondary-Node (“ADP not running”) via command “reboot”
2. Wait until Secondary-Node is up and running again
3. Reboot Primary-Node (“ADP is running”) via command “reboot”
4. Wait until Primary-Node is up and running again
b.) Separated Duplex system
1. Reboot Secondary-Node (“ADP not running”) via command “reboot”
2. Wait until Secondary-Node is up and running again
3. Reboot Primary-Node (“ADP is running”) via command “reboot”
4. Wait until Primary-Node is up and running again
5. Reboot Quorum-Node via command “reboot”
Next actions @development
SEN LC Development/GVS is investigating a possible solution/workaround with less Service impact, i.e. to fix the issue without requiring a system “reboot”.
“Leap second handling” on SUSE/NOVELL Enterprise Linux (SLES11) based Large Converged products
Message Body:
SEN LC Products based on SUSE/NOVELL Enterprise Linux (SLES11) can be affected by “leap second handling”.
There is a chance that the kernel leap second notice may cause an overload with a blockage (high CPU load/utilization) and a wide range of effects could be noticed, this could be everything from licensing issues (CLA) to an unstable system, e.g. RMX restarts/reloads or LINUX reboot.
How to verify if your system is affected?
To verify if your system is affected please check the /var/log/messages regarding the following entry:
Jul 1 01:59:59 linux kernel: [103264.364034] Clock: inserting leap second 23:59:60 UTC
If this message is present and your system/kernel has not been rebooted since this event we recommend to plan a service window to perform a system reboot via LINUX command: “reboot”.
Important Note: LINUX systems should be restarted via controlled reboot therefore POWER OFF must be avoided as usual.
Attention: In case your system was affected and it performed a complete LINUX reboot after 01.07.2012/23:59 (UTC) then no further action is needed because the reboot itself resolved the instability issues.
This behavior reflects the current status of analysis of GVS/ Development which is still ongoing.
Which products may be affected?
Following products are using affected LINUX kernel versions and could be impacted when a NTP Service is in operation:
HiPath 4000 V6 systems
Softgate V5
Softgate V6
How to perform recommended system reboot with HiPath4000 V6 Duplex deployment?
a.) Duplex system
1. Reboot Secondary-Node (“ADP not running”) via command “reboot”
2. Wait until Secondary-Node is up and running again
3. Reboot Primary-Node (“ADP is running”) via command “reboot”
4. Wait until Primary-Node is up and running again
b.) Separated Duplex system
1. Reboot Secondary-Node (“ADP not running”) via command “reboot”
2. Wait until Secondary-Node is up and running again
3. Reboot Primary-Node (“ADP is running”) via command “reboot”
4. Wait until Primary-Node is up and running again
5. Reboot Quorum-Node via command “reboot”
Next actions @development
SEN LC Development/GVS is investigating a possible solution/workaround with less Service impact, i.e. to fix the issue without requiring a system “reboot”.