Retest the fixed bugs of #1 on libtacky develop (batch 5) #24
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Retest of the bugs marked fixed in #1 (batch 5, headless, #1's method), and a check that the fixes broke nothing nearby. Results go here; #1 keeps its original evidence and gets a link back once a row is confirmed. The #9 retest with the GUI VMs (batch 6) is #25.
Versions
49b4c635, 2026-11-10); develop has 5 more commits toward 1.2.8 (room invites,65a31465)a9d303a2, libtacky v1.2.7); develop has 3 more (XEP-0245/medrawn as an action, b187c257 citing this repo's batch 3; the undecryptable line)master@b9e675dd; slixmpp (#1)Batch 5: #1 retest (headless, #1's method)
libtacky under test: develop (newer than v1.2.7), built as
tackyd-json-develop-<sha>; the exact commit is noted here when the batch starts.The interop runs of #1: libtacky's tackyd-json against slixmpp on the three servers, proof in
client.log/client-events.log/peer-slixmpp.log/server.log, branchvm/batch-5.All #1 bugs
Every bug of #1 (its Contents table), with the status as of libtacky develop
65a31465(no newer commits or releases, and no new comments on #1, at 2026-11-10 ~17:00 UTC). Retest yes = fixed or partly fixed upstream (67 of 73); the rest stay listed, not run. Result: ⏳ pending, then ✅ fixed / 🟡 partly / ❌ still there, with the run.65a31465)geoloc stopovertaken by a stale republishexpr(Inf)parse_urisplits a decoded?(link spoofing), accepts portsby=own account)<Ack>, XML console)fromthrows, message lost<close/>, 60 s stall (callback arity)notify <Notify><advanced/>dropped on rewrite<advanced/>kept)<delete/>stops sharing for goodexpr(1E…→Inf)<file/>lacks<hash/>and<desc/>urn:ietf:params:xml:ns:vcard-4.0urn:xmpp:eme:0;name=on listed EME nspubsub#notify_retractfalsemessage sendneither validates nor prepares JIDsaccount unregisternever answers on ejabberd; redial loop<message/>in urn:xmpp:forward:0 namespaceiap/happy_eyeballssettings never reach the connectionbgerror)adhoc listkeeps foreign command nodesthread_parent == threadaccepted<attachments>read as attachmentsget_invoicewithout disco<r/>unanswered ~50 min without declaring the stream dead<count>hiddenuntrust_certificatefor a registration domain fails<nick>accept_inviteleaves its autojoin bookmarkstanza-id by=(untested)Results
Batch 5: 199 runs, libtacky develop
65a31465(tackyd-json-develop-65a31465) against slixmpp on the three servers, PR #28 (vm/batch-5). Run verdicts: 187 pass, 9 fail, 3 na (peer or harness, each rerun).rega regression run;=0,iap,R2,live,mucsubthe part a run covers.<count>is checked on ejabberd and anotherim-server; Prosody's MAM<fin/>carries none. C30 can't be reached on ejabberd (it names the inviter by real JID); its regression run there passes. C17 R2 is read from runs with OMEMO on (ai, and the ej S2 run). Two fail verdicts are not these rows: ej C8/M16 run (C8 (b),notify get, ruled not a bug) and the ej M19 run (C7's Dino invite id, kept on purpose).<invalid-from/>0.94 s after the forged message; libtacky dropped the last stanza sent (iq 170), replayed the forged one on the next session, and lost legitimate iqs (MAM page, bundle fetch) (174303).Retest the fixed bugs of #1 and #9 on libtacky v1.2.7 / anotherim-js 0.8.0to Retest the fixed bugs of #1 on libtacky develop (batch 5)