[Chtc-users] All CHTC servers to be rebooted after 3pm for security vulnerability

[chtc-users@xxxxxxxxxxx]

Greetings CHTC Users,

Due to a recently announced security vulnerability in an essential piece of server software (glibc), we will need to reboot ALL CHTC servers as soon as possible. While we would prefer to give you much more notice in these situations, the nature of this vulnerability is such that we need to act quickly to maintain the security of our compute systems for all users.

ALL servers managed by CHTC will be rebooted sometime shortly after 3pm today (Central Time).

For the HPC Cluster

• The head nodes will be down and inaccessible for a brief period after 3pm.
• All running jobs will be interrupted sometime after 3pm and will likely need to be resubmitted.Â
• Queued jobs that have not started running yet *should* remain in the queue, but will not startÂrunning until after the reboot of all cluster servers.
  

For the HTC System (via HTCondor and submit nodes)

• CHTC-owned submit servers (including group servers run by CHTC) will be down and inaccessible for a brief period after 3pm.
• All queued jobs will remain in the queue, and any interrupted jobs will simply be rerun by HTCondor without you having to resubmit them.Â
• All jobs running on CHTC execute servers will be interrupted when those servers are rebooted, and jobs running on non-CHTC servers (via Flocking or Glidein) may be interrupted as well.
  

We do apologize for the inconvenience, but assure you that rebooting servers today is necessary. Compute centers on other campuses are taking similar immediate action.

If you have any specific questions, please send them to chtc@xxxxxxxxxxx, as always.

Thank you,

Your CHTC Team