@abh3 From my point of view, supporting only >=RH7 is fine. I'm not sure if there really are sites who run RH6 on their transfer nodes / file servers still. Of course, it might be that they have just installed them together with the batch nodes, and never upgraded (it happens...). But we can also argue CentOS6 is officially out of support in half a year anyways, and indeed for machines exposed to the internet, it's already now not a good diea to run it.

On our end (but we are only a single Tier3, so it's hard to extrapolate from my point of view), we are now using containers for any batch jobs and stay up to date on newest RH7 releases. RH8 is still a no-go for us (mainly, since we need CephFS which only supports this in the very latest versions and rely on the WLCG repositories for VOMS setup who are not ready yet). However, once those issues are sorted out, we want to upgrade as soon as possible. I think many sites in high energy physics are trying to decouple OS upgrades from the experiments' software stack using containers now, so this is a growing trend.

So in short, my opinion is to go >= RH7 only, since it would eat a lot of developer energies to still support the dead RH6 platform. Restarting via a cronjob is still an option if somebody really runs RH6 until it goes EoL anyways. It could be interpreted as a "deprecation warning" ;-).

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.

Use REPLY-ALL to reply to list

To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1