fedora-infrastructure

#12947 Unresolvable log01 in AWS
Closed: Fixed by msuchy. Opened by msuchy.

Closed: Fixed
Reopen Issue

In Copr we see in journal that "cannot resolve log01'.
For more details see https://github.com/fedora-copr/copr/issues/3377
The problem seems to be in this line in ansible.git:

roles/base/files/rsyslog/rsyslog-audit.conf.default:local6.* @@log01:514

When I try to resolve log01 on batcave01 it works, because /etc/resolv.conf has

search rdu3.fedoraproject.org fedoraproject.org

and it correctly tries log01.rdu3.fedoraproject.org.

Machine in AWS (definitely not Copr) does not have rdu3.fedoraproject.org in resolve.conf. But even when I try to add it, it does not work. Likely because:

# dig @172.30.0.2 log01.rdu3.fedoraproject.org
; <<>> DiG 9.18.41 <<>> @172.30.0.2 log01.rdu3.fedoraproject.org
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NXDOMAIN, id: 11906
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 1
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;log01.rdu3.fedoraproject.org.  IN      A
;; AUTHORITY SECTION:
fedoraproject.org.      167     IN      SOA     ns01.fedoraproject.org. hostmaster.fedoraproject.org. 2763754720 3600 600 2419200 86400
;; Query time: 15 msec
;; SERVER: 172.30.0.2#53(172.30.0.2) (UDP)
;; WHEN: Sat Nov 29 20:32:15 UTC 2025
;; MSG SIZE  rcvd: 109

I.e. log01.rdu3.fedoraproject.org is not resolvable from AWS.

Metadata Update from @zlopez:
- Issue priority set to: Waiting on Assignee (was: Needs Review)
- Issue tagged with: aws, high-gain, medium-trouble

Yes, this is the case. (and always has been. ;)

I suspect the rsyslog templates / ansible changed somewhere where it was giving copr hosts a config without log01?

So, our options are:

  1. Just fix ansible to not try and configure log01 on copr hosts.

  2. If we want copr hosts to log to log01, we probibly would need to add them to our vpn. I'm not sure we want to do this, but I suppose we could.

Thoughts? Do you need/want central logging there? Or not needed?

I am fine with option one. This will likely affect anything in AWS. Not just copr.

We are excluding the 'copr' group:

https://pagure.io/fedora-infra/ansible/blob/main/f/roles/base/tasks/main.yml#_461

I guess this should be 'copr_aws' group now?

right.
Thank you.

Except I misread it...

when: not inventory_hostname.startswith(('mirrorlist','copr'))

so it should exclude copr* hostnames... which should work. ;(
Could it be it was deployed wrong a while back and then fixed after? Can you try removing that file and re-running and see if it adds it back or not?

Hmm still does not work. :(

[root@copr-fe ~][PROD]# cat /etc/rsyslog.d/rsyslog-log01.conf
cat: /etc/rsyslog.d/rsyslog-log01.conf: No such file or directory
[root@copr-fe ~][PROD]# journalctl |grep log01 |tail
...
Jan 13 06:54:38 copr-fe.aws.fedoraproject.org rsyslogd[1214]: cannot resolve hostname 'log01' [v8.2508.0-1.fc43 try https://www.rsyslog.com/e/2027 ]
[root@copr-fe ~][PROD]# date
Tue Jan 13 07:04:29 AM UTC 2026
[root@copr-fe ~][PROD]# cat /etc/rsyslog.d/rsyslog-log01.conf
cat: /etc/rsyslog.d/rsyslog-log01.conf: No such file or directory
[root@copr-fe ~][PROD]# systemctl restart rsyslog
[root@copr-fe ~][PROD]# journalctl |grep log01 |tail
...
Jan 13 07:16:54 copr-fe.aws.fedoraproject.org rsyslogd[3606025]: cannot resolve hostname 'log01' [v8.2508.0-1.fc43 try https://www.rsyslog.com/e/2027 ]
Jan 13 07:18:04 copr-fe.aws.fedoraproject.org rsyslogd[3606025]: cannot resolve hostname 'log01' [v8.2508.0-1.fc43 try https://www.rsyslog.com/e/2027 ]
[root@copr-fe ~][PROD]#

Metadata Update from @kevin:
- Issue assigned to kevin

ah, it's the audit one.

Should be fixed in 8caea853adbef55a5aac680c8240414f3f98c666

You can manually delete /etc/rsyslog.d/rsyslog-audit.conf or just re-run the playbook to fix it.

Sorry for the hassle. I guess we had copr machines in the 'cloud' datacenter before, but it's in 'aws' now, and that needed an override for this file.

Metadata Update from @kevin:
- Issue close_status updated to: Fixed
- Issue status updated to: Closed (was: Open)

That works! Thank you.

Metadata Update from @msuchy:
- Issue status updated to: Open (was: Closed)

Metadata Update from @msuchy:
- Issue close_status updated to: Fixed
- Issue status updated to: Closed (was: Open)
Metadata

aws high-gain medium-trouble

None
None
Normal
Static archive. pagure.io was sunset in 2026; this is a read-only snapshot.
Red Hat Logo

Fedora is sponsored by Red Hat. Learn more about the relationship between Red Hat and Fedora »

© Red Hat, Inc. and others.