From Fedora Project Wiki

< FWN‎ | Beats

Line 8: Line 8:
Contributing Writer:  HuzaifaSidhpurwala
Contributing Writer:  HuzaifaSidhpurwala


== So Everyone is Aware ==
=== Some noteworty praise ===


Mike McGrath wrote for @fedora-infrastructure[1]
Paul W. Frields writes for fedora-infrastructure-list [1]


This is the first notice which came out to the community to inform them that there will be outages and a lot of the servers are being rebuilt. Mike pointed to the mail on @fedora-announce[2]
Paul forwarded a mail [2] send by Tim Burke, who is the Director of Linux Development inside Red Hat, praising the efforts of fedorans who rose to the occasion to bring things back on track after the recent incidents in Fedora infrastructure.


[1] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00108.html
[1] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00149.html
[2] https://www.redhat.com/archives/fedora-devel-list/2008-August/msg01023.html


[2] http://www.redhat.com/archives/fedora-announce-list/2008-August/msg00008.html
=== Maintaining a partial cvs workarea ===
 
Axel Thimm writes for fedora-infrastructure-list [3]
 
Axel described how he was keeping a partial check-out of packages, ie the ones which he was maintaining. Now he would like to be able to cvs up and have all updates flow in, but if he does do so cvs will want to get all other thousand packages in. He is currently using a for loop with pushd/popd, but this process is extremely slow. Axel asked if there was a better way of doing this?
 
[3] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00156.html
 
=== rawhide, /mnt/koji and /pub/fedora ===
 
Jesse Keating writes for fedora-infrastructure-list [4]
 
Jesse created a user "masher" to have the ability to write to /mnt/koji/mash/ but not any of the other koji space.  This is useful to prevent too much damage from a horribly wrong
rawhide compose.  To make things easier in the rawhide compose configs, they decided to run the cron/scripts as the masher user.  This is also good because it means things run unprivileged.  However he ran into a snag.  They have another user, 'ftpsync' that has write access to /pub/fedora/.  Previously the rawhide script was ran as root, and thus it was no problem to su ftpsync for the rsync calls.  The masher user does not possess the capability of doing this.
 
 
[4] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00174.html

Revision as of 03:40, 1 September 2008

Infrastructure

This section contains the discussion happening on the fedora-infrastructure-list

http://fedoraproject.org/wiki/Infrastructure

Contributing Writer: HuzaifaSidhpurwala

Some noteworty praise

Paul W. Frields writes for fedora-infrastructure-list [1]

Paul forwarded a mail [2] send by Tim Burke, who is the Director of Linux Development inside Red Hat, praising the efforts of fedorans who rose to the occasion to bring things back on track after the recent incidents in Fedora infrastructure.

[1] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00149.html

[2] https://www.redhat.com/archives/fedora-devel-list/2008-August/msg01023.html

Maintaining a partial cvs workarea

Axel Thimm writes for fedora-infrastructure-list [3]

Axel described how he was keeping a partial check-out of packages, ie the ones which he was maintaining. Now he would like to be able to cvs up and have all updates flow in, but if he does do so cvs will want to get all other thousand packages in. He is currently using a for loop with pushd/popd, but this process is extremely slow. Axel asked if there was a better way of doing this?

[3] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00156.html

rawhide, /mnt/koji and /pub/fedora

Jesse Keating writes for fedora-infrastructure-list [4]

Jesse created a user "masher" to have the ability to write to /mnt/koji/mash/ but not any of the other koji space. This is useful to prevent too much damage from a horribly wrong rawhide compose. To make things easier in the rawhide compose configs, they decided to run the cron/scripts as the masher user. This is also good because it means things run unprivileged. However he ran into a snag. They have another user, 'ftpsync' that has write access to /pub/fedora/. Previously the rawhide script was ran as root, and thus it was no problem to su ftpsync for the rsync calls. The masher user does not possess the capability of doing this.


[4] https://www.redhat.com/archives/fedora-infrastructure-list/2008-August/msg00174.html