I suspended the project and installed the update. After reboot, BOINC starts, and the activity monitor shows that sulphur is running, but BOINC cannot open any GUI windows. (Also posted to the BOINC manager message board at http://boinc.berkeley.edu/dev/forum_forum.php?id=1] .)

Quitting BOINC and re-starting it fixes the manager GUI problem. Now wondering why suspended work does not resume...

---

We can\'t see your computer because it\'s hidden, so you\'ll have to tell us a few things.

***************

Re Suspending:
There are 3, in different places, each giving a different level of control.
1) In the menu; This suspends BOINC.
2) In the Projects tab; This suspends whichever project\'s name is highlighted.
3) In the work tab; This suspends whichever model is highlighted.

If you want to Suspend before Exiting, the one to use is 1). I got into the habit of doing this over a year ago, and, while it is apparently no longer necessary with the latest version of BOINC, I like to shut down programs gently, rather than just \"pull the rug out from under them\".

If you use 3), BOINC soon thinks \"The science app isn\'t doing anything; It must have run out of work, so I\'ll get some more\". And you end up with 2 models.
So to use this one it is necessary to add an extra step first: In the projects tab, highlight the project whose model is going to be Suspended, and then click \"No new work\".

Which one did you use?

**************

What is the exact wording of the status message in the Work tab?

**************

Please copy the last 10-20 lines from the Messages tab, and paste them here.

Ah! I used the \"Projects\" suspend, not the menu. Perhaps it is still necessary to use the menu suspend. Anyway, I exitted BOINC and re-started it; then the GUI could open but the project \"Resume\" button did not take affect. Finally I tried the menu buttons. The second job has been aborted and the original job is running again.

Uhh, the activity monitor shows that the second (aborted) job is still in memory but using no CPU time. It\'s no longer visible from inside BOINC and so no obvious way to remove it from within BOINC. Is it safe to select and kill it using the activity monitor?

---

The activity monitor shows that the second (aborted) job is still in memory but using no CPU time. It\'s no longer visible from inside BOINC and so no obvious way to remove it from within BOINC. Is it safe to select and kill it using the activity monitor?

Maybe no longer of interest but here are the last messages:

Thu Mar 2 21:04:30 2006||Starting BOINC client version 5.2.13 for powerpc-apple-darwin
Thu Mar 2 21:04:30 2006||libcurl/7.14.0 OpenSSL/0.9.7i zlib/1.1.4
Thu Mar 2 21:04:30 2006||Data directory: /Library/Application Support/BOINC Data
Thu Mar 2 21:04:30 2006||Processor: 1 Power Macintosh PowerMac6,4
Thu Mar 2 21:04:30 2006||Memory: 768.00 MB physical, 0 bytes virtual
Thu Mar 2 21:04:30 2006||Disk: 74.41 GB total, 55.36 GB free
Thu Mar 2 21:04:30 2006|climateprediction.net|Computer ID: 387594; location: ; project prefs: default
Thu Mar 2 21:04:30 2006||General prefs: from climateprediction.net (last modified 2006-02-27 21:51:22)
Thu Mar 2 21:04:30 2006||General prefs: using your defaults
Thu Mar 2 21:04:30 2006|climateprediction.net|Deferring computation for result sulphur_imqy_200869290_0
Thu Mar 2 21:04:30 2006|climateprediction.net|Resuming computation for result sulphur_hpbn_100825971_1 using sulphur_cycle version 422
Thu Mar 2 22:12:17 2006||Suspending computation and network activity - user request
Thu Mar 2 22:12:17 2006|climateprediction.net|Pausing result sulphur_hpbn_100825971_1 (left in memory)
Thu Mar 2 22:13:20 2006||request_reschedule_cpus: project op
Thu Mar 2 22:13:46 2006||request_reschedule_cpus: result op
Thu Mar 2 22:15:34 2006|climateprediction.net|Unrecoverable error for result sulphur_hpbn_100825971_1 (aborted via GUI RPC)
Thu Mar 2 22:15:34 2006||request_reschedule_cpus: result op
Thu Mar 2 22:15:34 2006||request_reschedule_cpus: process exited
Thu Mar 2 22:15:34 2006|climateprediction.net|Computation for result sulphur_hpbn_100825971_1 finished
Thu Mar 2 22:15:55 2006||request_reschedule_cpus: result op
Thu Mar 2 22:15:56 2006||request_reschedule_cpus: result op
Thu Mar 2 22:15:59 2006||request_reschedule_cpus: result op
Thu Mar 2 22:16:01 2006||request_reschedule_cpus: result op
Fri Mar 3 06:25:55 2006||request_reschedule_cpus: result op
Fri Mar 3 06:25:58 2006||request_reschedule_cpus: result op
Fri Mar 3 06:26:11 2006||request_reschedule_cpus: result op
Fri Mar 3 06:26:40 2006||Resuming computation and network activity
Fri Mar 3 06:26:40 2006||request_reschedule_cpus: Resuming activities
Fri Mar 3 06:26:40 2006|climateprediction.net|Restarting result sulphur_imqy_200869290_0 using sulphur_cycle version 422
Fri Mar 3 06:26:40 2006|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
Fri Mar 3 06:26:40 2006|climateprediction.net|Reason: Requested by user
Fri Mar 3 06:26:40 2006|climateprediction.net|Reporting 1 results
Fri Mar 3 06:26:45 2006|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
Fri Mar 3 06:51:20 2006||request_reschedule_cpus: result op
Fri Mar 3 06:51:21 2006|climateprediction.net|Pausing result sulphur_imqy_200869290_0 (left in memory)
Fri Mar 3 06:51:22 2006||request_reschedule_cpus: result op
Fri Mar 3 06:51:23 2006|climateprediction.net|Resuming result sulphur_imqy_200869290_0 using sulphur_cycle version 422
Fri Mar 3 06:51:47 2006||Suspending computation and network activity - user request
Fri Mar 3 06:51:47 2006|climateprediction.net|Pausing result sulphur_imqy_200869290_0 (left in memory)
Fri Mar 3 06:52:48 2006||Resuming computation and network activity
Fri Mar 3 06:52:48 2006||request_reschedule_cpus: Resuming activities
Fri Mar 3 06:52:48 2006|climateprediction.net|Resuming result sulphur_imqy_200869290_0 using sulphur_cycle version 422

---

The activity monitor shows that the second (aborted) job is still in memory but using no CPU time. It\'s no longer visible from inside BOINC and so no obvious way to remove it from within BOINC. Is it safe to select and kill it using the activity monitor?

Yea, verily; it was safe. The original job still runs.

---