I might be the only Fedora user in Korea. I noticed that updates got slow again. Visited the only mirror in Korea to see this.
fedora: sync begins at 2024-03-31 09:54:01 +0900
+ source=rsync://dl-iad03.fedoraproject.org/fedora-enchilada/
+ site=ftp.kaist.ac.kr
+ node=mirror.ftp.kaist.ac.kr
+ triggered=regularly
+ frequency=PT10M
+ timepast=P4DT12H32M56S
+ failures=12
+ ffts=fullfiletimelist-fedora
++ rsync --no-motd --dry-run --out-format=%n rsync://dl-iad03.fedoraproject.org/fedora-enchilada//fullfiletimelist-fedora ./data/fullfiletimelist-fedora
@ERROR: max connections (20) reached -- try again later
rsync error: error starting client-server protocol (code 5) at main.c(1863) [Receiver=3.2.7]
+ checkresult=
++ exitcode=5
++ set +x
fedora: sync failure at 2024-03-31 09:54:02 +0900
It’s annoying. Hate it when dnf chooses a painfully slow mirror from the neighbouring countries and I have to rerun the dnf to force it to choose another.
Judging by the hostname its using(dl-iad03.fedoraproject.org), I think ftp.kaist.ac.kr predates the current tiering system of mirrors because that master is not listed here.
https://fedoraproject.org/wiki/Infrastructure/Mirroring/Tiering
Just like the last time, it appears to be the problem at the master end, not the mirror end. Obviously, I’ll contact the admin but I wanted to do them a favor by posting this. What’s going on? Is the master running under the pump? Perhaps it’s time for the mirror to change the source from the master to Tier-1 mirrors?