Project homepage Mailing List  Warmcat.com  API Docs  Github Mirror 
    npro  
 Modern all-safe Rust Network Protocol library supporting h1, h2, h3, ws, wt sans-IO and with socket IO + tls
git clone https://npro.rs/repo/npro
 
root / assets / solaris-11.svg
Author[]google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.c...> 2025-09-17 14:32 UTC
Committer[]Andy Green <andy@warmcat.com> 2025-09-17 14:33 UTC
Treeb06b3caaa3fd2a331814e2c5ce0b59371e938dc3   Raw Patch
 
server: Add diagnostic logging for rebuild loop
server: Add diagnostic logging for rebuild loop

A bug is causing completed jobs to be immediately re-issued by the
server, resulting in an infinite loop. The root cause appears to be that
the parent 'event' is not being correctly transitioned to a 'completed'
state.

This patch adds detailed logging to the sais_set_task_state() function.
It logs the old and new states for both the task and the event, and the
counts of total, good, and bad tasks that are used to determine the
event's state.

This information will be used to diagnose why the event state is not
being set correctly.
diff --git a/src/server/s-task.c b/src/server/s-task.c index 28408c0..7aa06be 100644 --- a/src/server/s-task.c +++ b/src/server/s-task.c @@ -244,8 +244,9 @@ sais_set_task_state(struct vhd *vhd, const char *builder_name, * Decide how to set the event state based on that */ - lwsl_notice("%s: ev %s, count %u, count_good %u, count_bad %u\n", - __func__, event_uuid, count, count_good, count_bad); + lwsl_notice("%s: ev %s, task %s, state %d -> %d, count %u, good %u, bad %u, oes %d\n", + __func__, event_uuid, task_uuid, task_ostate, state, + count, count_good, count_bad, (int)oes); sta = SAIES_BEING_BUILT;
Page fetched 0s ago, creation time: 1ms (vhost etag hits: 0%, cache hits: 0%)