You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ca-proxy: snif-certd, snif-apnsd and snif-syncd copy every line they log to syslog (facility daemon), as well as to their files in log/, at a priority that says what it is: warning for a peer down or a name not issued now, err for a failed issue step, a refused DNS update, a sync conflict or a judge with no answer. Until now they wrote to files that a host's log shipping does not read, so a cluster peer down, which pauses enrolment without a judge, was seen only on the node. SNIF_SYSLOG=0 in snif.conf keeps the files only. snif-apnsd, where it cannot write log/, logs to syslog only instead of stderr. snif-syncd also logs a status line, its peers, queues and judge: hourly while all is well, every 5 minutes while a peer is not up or a queue is stale, so a node whose status lines stop has stopped (doc/cluster.md 7.5). test/syslog.t covers it.