ack. Thanks for catching this.
I've made some adjustments in
https://github.com/haiku/infrastructure/commit/16cc8f
$ telnet rsync.haiku-os.org 12000
Trying 104.248.198.131...
Connected to rsync.haiku-os.org.
Escape character is '^]'.
@RSYNCD: 31.0
^]
telnet> q
Give it a try and let me know.
Thanks!
-- Alex
September 22, 2019 6:00 AM, "A L" <mail@xxxxxxxxxxxxxx> wrote:
Hello,
The rsync.haiku-os.org mirror seems to have goon offline since the server
move. At least I can no
longer sync from it.
Seems to be a CNAME chain too, not illegal, but discouraged since it uses
more DNS resources.
# drill rsync.haiku-os.org
;; ->>HEADER<<- opcode: QUERY, rcode: NOERROR, id: 1434
;; flags: qr rd ra ; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;; rsync.haiku-os.org. IN A
;; ANSWER SECTION:
rsync.haiku-os.org. 3556 IN CNAME ingress.haiku-os.org.
ingress.haiku-os.org. 3556 IN CNAME limerick.ams3.haiku-os.org.
limerick.ams3.haiku-os.org. 3556 IN A 104.248.198.131
Doing a telnet to port 12000 gives no rsync banner as usually either.
//Anders
On 2019-09-04 15:27, Alexander von Gluck IV wrote:
September 3, 2019 1:54 PM, "Alexander von Gluck IV" > <kallisti5@xxxxxxxxxxx
<kallisti5@xxxxxxxxxxx?to=%22Alexander%20von%20Gluck%20IV%22%20<kallisti5@xxxxxxxxxxx>>>
> wrote:
Good afternoon!
Just a heads up we will be attempting our server switch this
evening. (4-5 Hours away CDT time)
Pretty much everything will be impacted in some way.
The following changes are happening:
1) Docker compose -> Docker swarm (single node for now) (more
flexible long-term and we get secret storage)
2) online.net -> DigitalOcean
(solve the horrible randomly 1 MiB/s - 130 MiB/s Managed iSCSI
service that kill our services)
block storage is 300MiB/s+ throughout all testing.
3) France -> Amsterdam (Still within the EU, woot!)
4) nightly images Our server -> Wasabi s3 (cheaper storage)
This is going to be one of the more risky changes we've done.
The rollback plan is switching back to online.net
Wish us luck! I'll keep everyone updated in this thread.
-- Alex
Well! I sent this yesterday, but freelists has been down over the last > 24
hours.
The server migration is complete. We saw the core services down for > roughly
15 minutes, and various outages and breakages for ~ 2 hours while I > fixed
the
rough edges. (we also moved to docker swarm from docker compose)
Everything should be functional with the following known issues:
* Nightly images behind. I'm moving these to https://wasabi.com ;>
<https://wasabi.com> to save on
costs... should be synched up soon.
* Builders need updated to point new images wasabi
We still have various build system issues that need worked through, > but
things are
running *MUCH* better at DigitalOcean vs online.net.
Keep me posted if anyone sees any other major issues.
-- Alex
