Just curious, I like to keep up to date with the latest, I know its developmental, but with the constant changing of client versions does it have any adverse effects on the scheduling? I mena 4.42 starts "getting the hang of things" ie getting debt under control after running for a while... Does this get all messed up when you change versions?

Plus it would be nice to have an updated list of changes made between versions, granted it is only a small leap from 4.42 to 4.43, but something had to have changed??

---

Cheers

Gav



Nothing is 'fool-proof', someone will always invent a better fool!

It's really here. maybe we'll run out of numbers soon.

Remember it's developmental software and can do some strange things. Use at own risk, or stay with "official" release clients.

Just curious, I like to keep up to date with the latest, I know its developmental, but with the constant changing of client versions does it have any adverse effects on the scheduling? I mena 4.42 starts "getting the hang of things" ie getting debt under control after running for a while... Does this get all messed up when you change versions?

If you downgrades to v4.32 or earlier you loses the info, but upgrading shouldn't change anything when it comes to long-term debt and so on.


Only change since v4.42 is a bug-fix to wait 5 seconds after starting core client before tries to connect with manager.

There's also some mac-specific changes for when the mac-build is released.


As always, v4.43 is development-version with possible hidden bugs, so don't complain if your computer starts chasing the mouse around in your room...

I exited 4.42, installed 4.43 directly over the top with NO uninstall or reboot, and I GOT a "Connection Failed" box and blank page.

Which machine(s)? How fast is the machine?

What firewall package are you using?

Did you receive any messages from the firewall?

---

----- Rom
BOINC Development Team, U.C. Berkeley
My Blog

Winxphomesp1
xp2200+

I reboot just now. The taskbar B was unresponsive about 20 seconds, then up popped the "Connection Failed" box.

when I installed over 4.42 I was prompted by the Firewall for first the Boincmgr, and then immediately following this, the prompt for the CLI came up. This activity is much more like the older versions 4.25 4.32 etc.

Mcafee internet suite. firewall build 6.1.6144

if there's anything else, I'm here for you

tony

Installed and running Ok on 3 systems.

No 'Connect Fails' at all.
Even allowed me enough time to click 'Allow' on the one system I hadn't already configured ZoneAlarm for 'Changes Often'.

---

OK Rom, I'm on my Work computer:
Winxp home sp2
same firewall
P4 1.8

I installed 4.43 directly over 4.42, got prompted by the firewall for Boinc manager, but not the CLI. The boinc manager opened and looked good. I checked the Messages tab and couldn't find the "version change detected" line. I reboot. Everything came up fine Boinc Manager opened, I didn't get a "connection failed" message, but I also haven't seen the firewall screen giving access to the CLI.

this if from my log:

5/19/2005 3:58:58 PM|SETI@home|Resuming computation for result 31dc04aa.14083.24178.179816.144_0 using setiathome version 4.09
5/19/2005 3:58:58 PM||Resuming computation and network activity
5/19/2005 3:58:58 PM|ProteinPredictorAtHome|Deferring communication with project for 1 days, 16 hours, 28 minutes, and 37 seconds
5/19/2005 3:58:58 PM|LHC@home|Deferring communication with project for 6 hours, 54 minutes, and 24 seconds
5/19/2005 3:58:59 PM||Can't resolve hostname [predictor.scripps.edu] valid name, no data record of requested type
5/19/2005 3:58:59 PM|ProteinPredictorAtHome|Couldn't start upload of h0007A_1_33537_4_0
5/19/2005 3:58:59 PM|ProteinPredictorAtHome|URL http://predictor.scripps.edu/predictor_cgi/file_upload_handler: error -113
5/19/2005 3:58:59 PM|ProteinPredictorAtHome|Backing off 1 hours, 25 minutes, and 53 seconds on upload of file h0007A_1_33537_4_0
5/19/2005 3:58:59 PM||schedule_cpus: must schedule
5/19/2005 4:00:58 PM||Can't resolve hostname [predictor.scripps.edu] valid name, no data record of requested type
5/19/2005 4:00:58 PM|ProteinPredictorAtHome|Couldn't start upload of h0008A_1_73697_4_1
5/19/2005 4:00:58 PM|ProteinPredictorAtHome|URL http://predictor.scripps.edu/predictor_cgi/file_upload_handler: error -113
5/19/2005 4:00:58 PM|ProteinPredictorAtHome|Backing off 1 hours, 14 minutes, and 56 seconds on upload of file h0008A_1_73697_4_1
5/19/2005 4:01:24 PM||Can't resolve hostname [predictor.scripps.edu] valid name, no data record of requested type
5/19/2005 4:01:24 PM|ProteinPredictorAtHome|Couldn't start upload of h0008A_1_73715_4_2
5/19/2005 4:01:24 PM|ProteinPredictorAtHome|URL http://predictor.scripps.edu/predictor_cgi/file_upload_handler: error -113
5/19/2005 4:01:24 PM|ProteinPredictorAtHome|Backing off 3 hours, 1 minutes, and 58 seconds on upload of file h0008A_1_73715_4_2

I signed online and then when this happened, I got the request for access.

5/19/2005 4:04:44 PM|ProteinPredictorAtHome|Started upload of h0008A_1_73715_4_1
5/19/2005 4:04:49 PM|ProteinPredictorAtHome|Finished upload of h0008A_1_73715_4_1
5/19/2005 4:04:49 PM|ProteinPredictorAtHome|Throughput 10011 bytes/sec
5/19/2005 4:06:24 PM|ProteinPredictorAtHome|Started upload of h0007A_1_30205_4_0
5/19/2005 4:06:26 PM|ProteinPredictorAtHome|Finished upload of h0007A_1_30205_4_0



I still haven't seen a connection failed for this puter after reboot.

tony

Just thought I'd give it a try on a Linux machine here. It's the worst I've ever tried. :( Doesn't do anything useful at all. It trashed all the existing WUs with a message about the files being twice the size of the known universe ...or something to that effect ;) and couldn't contact the scheduler for more. Ho hum.

---

Be lucky

Neil

Update. On the second reboot (a hard restart) of my xp2200 I haven't gotten the connection failed message.

---

update on the P4 1.8:

I did a hard boot and when it opened, I got the "Connection failed" message.

It seems like holding off a connection attempt until everything is loaded might be the fix. If you remember, I didn't get a error message with the soft boot where the scheduler didn't try a connection for 5 minutes after loading.

I don't know

tony

Just installed on all 4 of my "work" machines (the good ones). No problems. Three WinXP-SP2, one Win2K-SP4.

---

update on the P4 1.8

I did a soft boot (restart). I got two connection failed messages, one right after the other, see log below:

5/19/2005 4:33:34 PM||Starting BOINC client version 4.43 for windows_intelx86
5/19/2005 4:33:34 PM||Data directory: C:\Program Files\BOINC
5/19/2005 4:33:35 PM|Einstein@Home|Computer ID: 4681; location: ; project prefs: default
5/19/2005 4:33:35 PM|LHC@home|Computer ID: 21958; location: home; project prefs: default
5/19/2005 4:33:35 PM|ProteinPredictorAtHome|Computer ID: 45681; location: work; project prefs: default
5/19/2005 4:33:35 PM|SETI@home|Computer ID: 407818; location: work; project prefs: default
5/19/2005 4:33:35 PM||General prefs: from ProteinPredictorAtHome (last modified 2005-05-19 04:20:20)
5/19/2005 4:33:35 PM||General prefs: using separate prefs for work
5/19/2005 4:33:35 PM||Remote control not allowed; using loopback address
5/19/2005 4:33:35 PM|SETI@home|Resuming computation for result 31dc04aa.14083.24178.179816.144_0 using setiathome version 4.09
5/19/2005 4:33:35 PM|ProteinPredictorAtHome|Deferring communication with project for 1 days, 15 hours, 54 minutes, and 0 seconds
5/19/2005 4:33:35 PM|LHC@home|Deferring communication with project for 6 hours, 19 minutes, and 47 seconds
5/19/2005 4:33:35 PM||schedule_cpus: must schedule

update on the xp2200+

I did a soft boot, got two connection failed messages, one after the other. see log below:

5/19/2005 4:48:21 PM||Starting BOINC client version 4.43 for windows_intelx86
5/19/2005 4:48:21 PM||Data directory: C:Program FilesBOINC
5/19/2005 4:48:24 PM|Einstein@Home|Computer ID: 9495; location: ; project prefs: default
5/19/2005 4:48:24 PM|LHC@home|Computer ID: 22086; location: home; project prefs: default
5/19/2005 4:48:24 PM|ProteinPredictorAtHome|Computer ID: 73452; location: home; project prefs: default
5/19/2005 4:48:24 PM|SETI@home|Computer ID: 524220; location: home; project prefs: default
5/19/2005 4:48:24 PM||General prefs: from ProteinPredictorAtHome (last modified 2005-05-19 04:20:20)
5/19/2005 4:48:24 PM||General prefs: no separate prefs for home; using your defaults
5/19/2005 4:48:26 PM||Remote control not allowed; using loopback address
5/19/2005 4:48:33 PM|SETI@home|Resuming computation for result 31dc04aa.14083.15872.265890.78_1 using setiathome version 4.09
5/19/2005 4:48:34 PM||Suspending network activity - user request
5/19/2005 4:48:34 PM||schedule_cpus: must schedule
5/19/2005 4:49:48 PM|SETI@home|Result 31dc04aa.14083.15872.265890.78_1 exited with zero status but no 'finished' file
5/19/2005 4:49:48 PM|SETI@home|If this happens repeatedly you may need to reset the project.
5/19/2005 4:49:48 PM||request_reschedule_cpus: process exited
5/19/2005 4:49:48 PM||schedule_cpus: must schedule
5/19/2005 4:49:49 PM|SETI@home|Restarting result 31dc04aa.14083.15872.265890.78_1 using setiathome version 4.09

when I originally planned the reboot, I figured I would try disabling network access to see if that changed things. Instead it allowed network access until it loaded then turned network access off (see log above). hmm

For the first time in quite a while I did not get the "Connection Error" on either machine. But I did notice, my Einstien WU, Exited and said "if it continues to happen''''etc.

Rocky

---

www.boincsynergy.com

For the first time in quite a while I did not get the "Connection Error" on either machine. But I did notice, my Einstien WU, Exited and said "if it continues to happen''''etc.

Rocky

rocky, check your messages, do you see the version change detected line?

For the first time in quite a while I did not get the "Connection Error" on either machine. But I did notice, my Einstien WU, Exited and said "if it continues to happen''''etc.

Rocky

rocky, check your messages, do you see the version change detected line?

Yes..From 4.42 to 4.43

---

www.boincsynergy.com

For the first time in quite a while I did not get the "Connection Error" on either machine.
Rocky

Yes Rocky, same pleasure here. This is a bug i'm very happy to see squashed.
Unhappy to see mmciastro's problems. Time will tell if others experience this.

But I did notice, my Einstien WU, Exited and said "if it continues to happen''''etc.
Rocky

Have seen this before, most often exit and restart BOINC and all is good.

---

I have installed the 4.43 version. It deleted all my WU's, luckly not the ones that are waiting for upload. I have 21 WU saved in other location, I have moved them in the application folder but still it doesn't see them. I am running the "optimized" client...
If somebody have an ideea how to recuperate those units, please advise. I am sorry to trash them.
PS: My uploading/downloading is not working. I will reboot.

---