[10:45:50] I tried to reproduce the failing ipv6 ping probes from my vm in ams, but couldn't [10:46:02] also tried in both directions (from the bastions) [10:46:15] has anybody else tried by any chance? [11:18:51] (ams->sfo is for me via zayo, don't see major problems) [11:20:00] let's keep it monitored and see during the next hours, the number of failed probes seems not that alarming now [11:20:07] (only for sfo, small number) [11:20:29] before it was in most of the pops, but only ipv6 afaik [11:20:34] anyway, ttl :) [11:24:14] elukey: seeing thr big drop on https://grafana.wikimedia.org/d/K1qm1j-Wz/ripe-atlas?orgId=1&var-datasource=eqiad%20prometheus%2Fops&var-target_site=All&var-ip_version=ipv6&var-country_code=All&var-asn=All&fullscreen&panelId=4 [11:24:47] I'd say it's either the ripe atlas constellation that is having an issue, or a major provider [15:25:16] wow there was a drop on ipv4 as well [15:53:20] XioNoX: ah ok! with "ripe atlas constellation" you mean the probes reporting to ripe? [15:53:28] (trying to understand) [15:54:25] yeah, monitoring glitch [15:54:52] might be worth asking ripe if they know something, but we don't have control over it anyway [15:55:08] having the number of hops as a metric would be useful [15:56:01] yep yep, difficult to judge.. I was interested in checking since a lot of our anchors were reporting failures, so I thought about a provider that we use in multiple places mayb [15:56:05] *maybe [15:56:10] but couldn't repro in any way [15:57:08] anyway, allo good, thanks for checking :) [15:59:23] thanks for looking into it on a weekend :) [16:12:21] I went looking to see if there was a drop in the frontend traffic graph around the same time (there wasn't) but then found several other inexplicable things [16:12:33] (nothing really worrying, just strange) [16:15:06] never go looking at dashboards unless you already know what you're looking for ;) [16:40:44] <_joe_> ^^