There has been a root CA expiration on Sep. 30th, so make sure your certificate trust chain is up to date. https://letsencrypt.org/docs/dst-root-ca-x3-expiration-september-2021/ Although given that the error is `connection reset by peer`, I'm not sure that this is the actual issue. Best regards, Moritz On Mon, Oct 4, 2021 at 9:09 AM Adrian Schröter wrote:

On Sonntag, 3. Oktober 2021, 01:25:01 CEST Hans-Peter Jansen wrote:

...

Hi,

harvesting all kinds of strange OBS issues here:

home:frispete:RemoteDesktop/freerdp$ osc ci Sending _servicedata Deleting freerdp-3.0.0~git.20211002T180031.8435aa608.tar.xz Sending freerdp.changes Sending freerdp.spec Sending freerdp-3.0.0~git.20211003T003236.342fb6500.tar.xz Transmitting file data .... Committed revision 86. SSL Error: (104, 'Connection reset by peer')

home:frispete:kernel/virtualbox$ osc up Failed to reach a server: [Errno 101] Network is unreachable

OBS is slow as hell, trashed some files on co, and even fail to build packages locally (guess, due to API server timeouts).

hm, i noticed this on friday as well. Increasing the SSL session cache help back then.

Dunno what was the situation on saturday, atm it looks fine for me.

but thanks for reporting! adrian

--

Adrian Schroeter Build Infrastructure Project Manager

SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany (HRB 247165, AG München), Geschäftsführer: Felix Imendörffer

-- Quobyte GmbH, Berlin, AG Charlottenburg HRB 149012 B, Felix Hupfeld, Bjoern Kolbeck -- Quobyte GmbH, Berlin, AG Charlottenburg HRB 149012 B, Dr. Felix Hupfeld, Dr. Bjoern Kolbeck

Hey, On 04.10.21 10:15, Andreas Schwab wrote:

On Okt 04 2021, Adrian Schröter wrote:

...

Dunno what was the situation on saturday, atm it looks fine for me.

It lasted until late Sunday evening.

We had multiple spiders (bing, yandex, unicom, sprinthost.ru) crawling 'expensive' routes like /package/view_file/... Every once in a while their frequencies align and then it becomes a bit too much. But in general we want them to index things of course. But it wasn't *that* bad, 95% of all requests still where answered faster than 256 milliseconds. We are looking into a more clever robots.txt now and some request throttling in the mid-term future.... Henne -- Henne Vogelsang http://www.opensuse.org Everybody has a plan, until they get hit. - Mike Tyson

Hey, On 04.10.21 14:06, Adrian Schröter wrote:

On Montag, 4. Oktober 2021, 14:00:25 CEST Henne Vogelsang wrote:

...

On 04.10.21 10:15, Andreas Schwab wrote:

...

On Okt 04 2021, Adrian Schröter wrote:

...

Dunno what was the situation on saturday, atm it looks fine for me.

It lasted until late Sunday evening.

We had multiple spiders (bing, yandex, unicom, sprinthost.ru) crawling 'expensive' routes like

But it wasn't *that* bad, 95% of all requests still where answered faster than 256 milliseconds.

that was not the scenario we spoke about. We had > 6 seconds just for the SSL handshake.

The handshake was one issue we talked about yes.

this is independend of the used route.

Sure, as usual with an instance that serves 25 requests per second, there are multiple things going on at the same time :-) This was going on Friday evening to Sunday night. But some of them are still at it right now. Henne -- Henne Vogelsang http://www.opensuse.org Everybody has a plan, until they get hit. - Mike Tyson