Hello everyone, On Wed Oct 7, 2026 at 5:52 PM CEST, Jeremy Kerr wrote: > Hi Mehmet, > >> patchwork.ozlabs.org has been slow for us for some weeks, in a browser >> and through the REST API, and requests often time out. > > Particularly over the last week or so there have been some increased > loads, this tends to be the ongoing whack-a-mole situation with scraper > workloads.` > > Somewhat coincidentally, this tends to happen while I am travelling. I > have scheduled some time next week to tweak the anubis settings and > connection limits, when I am back at the desk. > >> Is this a known load problem on the machine, and is there anything >> users can do to lighten it, for example in how often they poll the >> API? > > Legitimate web- and API- uses are only a fraction of the patchwork > workload; the vast majority of problematic load is crawlers. There have > been some cases of high-load actual-user API polling in the past, but > that has been fairly easy to identify and resolve.
In August I posted a series of patches meant to optimize queries generated when someone browses the list of patches. See: https://lists.ozlabs.org/pipermail/patchwork/2026-August/007520.html While it won't solve the problem entirely (anubis is necessary), maybe it could help reduce load on the servers? I could try backporting it to an older version if that helps. Regards, Franek Stachura _______________________________________________ Patchwork mailing list [email protected] https://lists.ozlabs.org/listinfo/patchwork
