NOTE
If your issue is for security or deals with sensitive info please mark it as private using the checkbox below.
I'm out of disk quota on fedorapeople.org; du -h says I'm using about 2.3G (covering many years of work on Fedora, upstream Python, and upstream GCC), and quota says:
Disk quotas for user dmalcolm (uid 1494201425): Filesystem blocks quota limit grace files quota limit grace /dev/mapper/vg_vm-srv 2382208* 2097152 2097152 none 57007 0 0
I'm attempting to use mass prebuild to do early test rebuilds of Fedora with GCC 15 to identify compat problems, but am having trouble getting GCC to build (it takes ~30 hours per build, and I keep running into failures).
I have scratch builds from jakub in Koji; and have built per-arch repos from them. For mass-prebuild I need a public-facing repo. For all archs, this is 5.5G of rpms to upload.
How much of an ask is it for me to have (say) 6G extra quota for (say) the next 2 months?
Alternatively, is there a way to turn a koji scratch build into a public-facing dnf repo (I've scripted this, but I don't have the disk space to host it)
If this is trivial, it would be great if this could be done today (2024/12/06). Otherwise I'll need to look at other solutions.
Thanks!
Metadata Update from @kevin: - Issue assigned to kevin
No problem at all.
I upped it to 6g. Happy to add more if you need it.
Metadata Update from @kevin: - Issue close_status updated to: Fixed with Explanation - Issue status updated to: Closed (was: Open)
Thanks. Unfortunately I meant my request to be to add 6G to my quota, rather than to set it to 6G, and my rsync failed:
rsync: [receiver] write failed on "/home/fedora/dmalcolm/public_html/gcc/gcc-15-mass-prebuild/s390x/gcc-objc++-15.0.0-0.2.fc42.s390x.rpm": Disk quota exceeded (122) rsync error: error in file IO (code 11) at receiver.c(378) [receiver=3.2.3] rsync: [sender] write error: Broken pipe (32)
I think my quota was ~2G, so a total of 8G might (I hope) be enough, but maybe 10G for safety's sake? Sorry about this.
Metadata Update from @dmalcolm: - Issue status updated to: Open (was: Closed)
Oh, sorry.
Done. Should be 10G now.
Let us know if that isn't enough...
Metadata Update from @kevin: - Issue close_status updated to: Fixed - Issue status updated to: Closed (was: Open)
Unfortunately it still seems to be failing:
$ rsync -avz . dmalcolm@fedorapeople.org:public_html/gcc/gcc-15-mass-prebuild/ sending incremental file list s390x/ s390x/gcc-objc++-15.0.0-0.2.fc42.s390x.rpm rsync: [receiver] write failed on "/home/fedora/dmalcolm/public_html/gcc/gcc-15-mass-prebuild/s390x/gcc-objc++-15.0.0-0.2.fc42.s390x.rpm": Disk quota exceeded (122) rsync error: error in file IO (code 11) at receiver.c(378) [receiver=3.2.3] rsync: [sender] write error: Broken pipe (32)
and logging in shows:
[dmalcolm@people01 ~][PROD]$ quota Disk quotas for user dmalcolm (uid 1494201425): Filesystem blocks quota limit grace files quota limit grace /dev/mapper/vg_vm-srv 6285720 6291456 6291456 57292 0 0
I completely failed to update your entry when I tried to bump it to 10g. ;(
Fixed and pushed.
Can you confirm it's ok now?
Thanks; seems to have worked this time:
sent 1,878,414,202 bytes received 3,222 bytes 1,636,965.08 bytes/sec total size is 5,879,618,778 speedup is 3.13
Metadata Update from @dmalcolm: - Issue close_status updated to: Fixed - Issue status updated to: Closed (was: Open)
Sorry, but something seems to have happened (permissions?) and https://dmalcolm.fedorapeople.org/ is now giving a 404
It possibly started when I did a restorecon.
I've been reading and re-reading https://fedoraproject.org/wiki/Infrastructure/fedorapeople.org#Common_answers but can't see the problem
$ ls -alZ ~ total 284 drwxr-xr-x. 7 dmalcolm dmalcolm unconfined_u:object_r:user_home_dir_t:s0 4096 Dec 17 23:07 . drwxr-xr-x. 6290 root root unconfined_u:object_r:user_home_dir_t:s0 155648 Nov 11 19:59 .. -rw-------. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 665 Dec 9 21:33 .bash_history -rw-r--r--. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 33 Oct 21 2008 .bash_logout -rw-r--r--. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 176 Oct 21 2008 .bash_profile -rw-r--r--. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 124 Oct 21 2008 .bashrc -rw-rw-r--. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 17760 Feb 28 2020 '-fdiagnostics-color=always' drwxrwxr-x. 3 dmalcolm dmalcolm unconfined_u:object_r:user_home_t:s0 35 Feb 4 2010 fedora.bak -rw-------. 1 dmalcolm dmalcolm unconfined_u:object_r:user_home_t:s0 20 Dec 17 23:07 .lesshst -rw-rw-r--. 1 dmalcolm dmalcolm unconfined_u:object_r:var_t:s0 65 Mar 23 2009 .planet drwxrwxr-x. 10 dmalcolm dmalcolm unconfined_u:object_r:git_user_content_t:s0 184 Jul 25 2012 public_git drwxrwxr-x. 13 dmalcolm dmalcolm unconfined_u:object_r:httpd_user_content_t:s0 4096 Nov 18 2021 public_html
Perhaps this caused the issue:
$ restorecon -v . Relabeled /home/fedora/dmalcolm from unconfined_u:object_r:var_t:s0 to unconfined_u:object_r:user_home_dir_t:s0
?
It's not you. Something seems to have broken $user.fedoarpeople.org... if you see fedorapeople.org/~username works fine. ;(
I am at a loss for why tho.
fixed.
Needed do downgrade httpd more...
Metadata Update from @phsmoura: - Issue priority set to: Waiting on Assignee (was: Needs Review) - Issue tagged with: low-gain, low-trouble, ops
Log in to comment on this ticket.