Opened 8 years ago

Closed 6 years ago

#5737 closed Bug (fixed)

Webui not available on daemon started with systemd

Reported by: manuel Owned by: mike.dld
Priority: High Milestone: 2.90
Component: Daemon Version: 2.83
Severity: Critical Keywords: webui deamon systemd


I guess this is tightly related to #2761. The problem in this ticket is, that the page displayed seems to be in the browser cache. Giving this bug appearence. Theunderling bug is that the webui is not online. transmisson-daemon seems to handle the online status of an interface as mandatory for the webui. The deamon on my machine (arch) is booted via a systemd sevice.

  1 [Unit] 
  2 Description=Transmission BitTorrent Daemon 
  5 [Service] 
  6 User=manuel 
  7 Type=notify 
  8 ExecStart=/usr/bin/transmission-daemon -f --log-debug 
  9 ExecReload=/bin/kill -s HUP $MAINPID 
 11 [Install] 

The Webui is not available after a fresh boot. A simple service restart solves the problem. The definition of "network is up" is ambigious (See: and I guess transmission-daemon needs a gateway adress to start the webui. Which is conceptually not necessary. I attached my journalctl logs of transmission daemon combined with dhcpcd to see when the network is up.

Please start the webui independant of the interface status.

Attachments (1)

log.log (11.2 KB) - added by manuel 8 years ago.
Journalcrtl since boot (dhcpcd and transmisson-daemon)

Download all attachments as: .zip

Change History (4)

Changed 8 years ago by manuel

Journalcrtl since boot (dhcpcd and transmisson-daemon)

comment:1 Changed 8 years ago by andyka

Problem: No web interface at system startup and is fixed by transmission-daemon service restart. Headless Ubuntu Linux 14.04.1 server (amd64 arch) (updated from the 12.04 LTS version), transmission at version 2.82 (2.82-1.1ubuntu3.1). Problem existed also before updating the distro with transmission at version 2.51 (2.51-0ubuntu1.4).

comment:2 Changed 6 years ago by mike.dld

  • Owner set to mike.dld
  • Status changed from new to assigned

comment:3 Changed 6 years ago by mike.dld

  • Milestone changed from None Set to 2.90
  • Resolution set to fixed
  • Status changed from assigned to closed

With r14663+, RPC server will retry the bind 10 times with increasing delays, for overall duration of about 2 minutes. If that doesn't help, error message will be written to log.

The link in description suggests the use of IP_FREEBIND, but that is 1) Linux-specific (althought there's also IP_BINDANY etc. on other platforms), and 2) not easily done with libevent's evhttp we currently use. If there're other ideas on how to fix this, I'm open to suggestions.

Note: See TracTickets for help on using tickets.