Super interesting, and very helpful. Thanks Paul. Luke
On Tue, Jan 26, 2021 at 9:37 AM Paul Gregg <pgregg+mai...@pgregg.com> wrote: > On Tue, Jan 26, 2021 at 03:52:49AM -0700, Luke via mailop wrote: > > Starting very precisely on January 6th, we started to see a number of > > previously reliable domains turn up with secondary MX records that look > > like these: > > > > cluster1a.us.messagelabs.com > > cluster1a.eu.messagelabs.com > > cluster1a.uk.messagelabs.com > > > > The domains with these secondary records have primary mx records that > look > > like these: > > > > cluster1.us.messagelabs.com > > cluster1.eu.messagelabs.com > > cluster1.uk.messagelabs.com > > > > Starting on January 6th, successful deliveries to these domains declined > > about 50%, but did not drop to zero. Engagement at these domains has > > remained steady. However we are now seeing *tons *of *421 Service > > Temporarily Unavailable *responses from the secondary MX records while > > still seeing successful deliveries to the primary MX record. > > > > Curious to hear if anyone else is seeing something similar or if anyone > > can help me understand what might be going on with these cluster1a* MX > > records. > > > > Here are a few example domains: > > > > funaisoken.co.jp > > cimb.com > > stockland.com.au > > travelex.com > > > > Would love to hear any thoughts or insights. > > > > Cheers, > > Luke > > Can't really help other than to say we don't see anything like that. > > Looked at logs for yesterday - precisely zero 421 Service Temporarily > Unavailable responses from cluster1a? (both). > Some for other clusters, e.g. cluster10a > > If stats helps, we delivered over 15,000 messages to cluster1.(us|eu|uk) - > but zero to cluster1a - didn't need to drop to lower prio mx > > PG >
_______________________________________________ mailop mailing list mailop@mailop.org https://list.mailop.org/listinfo/mailop
