here is what you said earlier that your setiathome.xml looked like:

/
/http://setiathome.berkeley.edu//
/xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/
/SETI@home/
/
/495/
/
/SETI@home classic/

This does not have a venue of "home" in it. If you're saying it now does than maybe you should shut down boinc and restart it to see if it's fixed now.

I have 3 account xml files, one for each project:

account predictor
account seti
account setibeta

are you sure you don't have more than ONE account xml that may be messed up?

tony

No, i do not have more than ONE account file with the ending .xml in my BOINC directory. ''blink''Maybe BOINC means there should be an other one (Astropulse).''/blink''

---

_\|/_
U r s

Check to make certain that the file name does not have a leading or trailing space. Check to make certain that the URl in the client_state.xml file does not have a leading or trailing space. Check that the URL in the client_state.xml exactly matches the URL portion of the relavant account_*.xml file.

---

BOINC WIKI

Check to make certain that the file name does not have a leading or trailing space.

No leading or trailing spaces.

Check to make certain that the URl in the client_state.xml file does not have a leading or trailing space.

Also no leading or trailing spaces found on all three hosts.

Check that the URL in the client_state.xml exactly matches the URL portion of the relavant account_*.xml file.

The URL is exactly the same in the client_state.xml and the sah account file.

I do not see, why this should be invalid.

And i would prefer this host to report and download. The deadline for the 32 wus is only 9 days away.

---

_\|/_
U r s

Does the user running BOINC have read and write permissions in the BOINC directory? Have the permissions been mangled somehow for these particular files? Does some other application have them open?

At this point, I am grasping at straws.

---

BOINC WIKI

JM7 he's from Germany. His grasp of the english language is much better than my grasp of German. I think I'm having trouble making myself clear enough. Earlier he posted what was in his setiathome.xml file and it doesn't show a venue of home. Later he stated it did. at the same time he kept thinking filename was ONLY the account name(that was my impression). If it does say "home" now, but didn't before, maybe it's fixed now, but he hasn't tried it. I don't know. It's late I'm going to bed.

Urs, good luck

sorry I couldn't help

tony

JM7 he's from Germany. His grasp of the english language is much better than my grasp of German. I think I'm having trouble making myself clear enough. Earlier he posted what was in his setiathome.xml file and it doesn't show a venue of home. Later he stated it did. at the same time he kept thinking filename was ONLY the account name(that was my impression). If it does say "home" now, but didn't before, maybe it's fixed now, but he hasn't tried it. I don't know. It's late I'm going to bed.

Urs, good luck

sorry I couldn't help

tony

Yes, well, my German is almost completely non-existent.

---

BOINC WIKI

The file name for the account should be exactly:
account_setiathome.berkeley.edu.xml

---

BOINC WIKI

maybe it's Boinc 4.46 problem, which should not be use except by the ones requested by Rom.

---

Fred
BOINC Alpha, BOINC Beta, LHC Alpha, Einstein Alpha

maybe it's Boinc 4.46 problem, which should not be use except by the ones requested by Rom.

Prime example of why the way things are going is a real cluster. The link to 4.46 was here for quite some time today, and I'm sure a few folks downloaded it before ROM posted that it was "pre-alpha".

There seems to be zero "release control" here, and no, I'm not going to offer assistance, as I have have no doubt I'd be ignored. The delevolopers must know better than me, but then again, a question about some random person using a pre-alpha has got me scratching my head....

Out of 100000 users, any link offering new SW (be it pre-alpha) will be used by some precent, and even if that's 1% or .5%, that's enough to confuses the picture! This post is just a case in point!

---

I think that Urs Echternacht was having the problem with 4.45. I think he gave 4.46 a try to see if it would solve his problem. When it didn't he switched back to 4.45 and is still having trouble reporting a number of work units.

---

Do you actually read the thread before jumping to conclusions? He Was and IS using 4.45, he tried 4.46 as a possible fix, but it didn't work. NOTE: as far as I know only Heffed and myself have been requested to use 4.46. Here's what Rom sent to another user asking about 4.46 in the dev list:

It is a pre-alpha drop.

I am having problems with my home phone configuration, so I asked a couple
of people whom I know are on dial-up to test a few changes for me.

I didn't make an announcement about it, but as always people are watching
everywhere... :)

LOL

----- Rom

tony

I chased through the code today and the error message is now correct,but won't be out till the next release. But the message shoud be: "Invalid account filename: account_setup" ...

I can't make heads nor tails of the C++ code, but my best guess is that there is a file named either "setup", or "account_setup" in the directory and that is causing this message. It is a benign warning ...

If someone better at reading C++ code wants to look it is in files.C ...

Anyway, that is as far as my research has taken me with that message ... which is summed up in the Wiki text for that message ... so youse all were reading it just as I was writing it ... :(

---

I chased through the code today and the error message is now correct,but won't be out till the next release. But the message shoud be: "Invalid account filename: account_setup" ...

I can't make heads nor tails of the C++ code, but my best guess is that there is a file named either "setup", or "account_setup" in the directory and that is causing this message. It is a benign warning ...

If someone better at reading C++ code wants to look it is in files.C ...

Anyway, that is as far as my research has taken me with that message ... which is summed up in the Wiki text for that message ... so youse all were reading it just as I was writing it ... :(

It would be my pleasure!
Only, I can't say what we are looking for is there (yet)!

I'll investigate!

---

Soo, well, i'm back again after a hot day in germany (29C in the shadow).
Let's summarize what happened here last night (night on my side of earth), before everybody starts to speculate:

Problem 1. (Main-problem)
''No schedulers responded'' is still the status quo on one of my hosts (hostid 997552). It has run out of work somewhen last week and is not reporting/downloading wus. There're 32wus waiting to be reported with deadlines of the 5th and 7th July.
Yesterday i tried to use CC4.46 in hope of a fix for the problem, but had no luck. I got a somewhat more verbose error message (500) during that test and reported this to Paul D. Buck, who forwarded the report to Rom Walton. After that i switched that host back to CC4.45 with my problem unsolved.

Another host (hostid 8255) had the same problem, but after a much to high benchmark result yesterday in the late evening, it started to report and download wus a little later. This host is running fine since. It's next connection will happen tuesday 28th June.


Problem 2.
On every startup of Boinc i get the warning message ''Invalid account file : account_setup'' on all of my hosts since i switched a month ago CC4.19 -> CC4.43+.
mmciastro has seen this message in one of my posts and pointed me to it. We tried to fix that last night but had no luck so far. We even did misunderstand one the other, sorry for that.
To clarify this issue i post my account file named ''account_setiathome.berkeley.edu.xml'' again in a more readable way and this time hopefully complete :
(the authenticator i replaced with some x's, 'lt'- and 'gt'-signs with a '-')

-account-
-master_url-http://setiathome.berkeley.edu/-/master_url-
-authenticator-xxxxxxxxxxxxxxxxxxxxxxxxxxxx-/authenticator-
-project_name-SETI@home-/project_name-
-host_venue-home-/host_venue-
-project_preferences-
-resource_share-495-/resource_share-
-project_specific-
-format_preset-SETI@home classic-/format_preset-
-text_style-Pillars-/text_style-
-graph_style-Rectangles-/graph_style-
-max_fps>30-/max_fps-
-max_cpu-50-/max_cpu-
-grow_time-10-/grow_time-
-hold_time-5-/hold_time-
-graph_alpha-0.7-/graph_alpha-
-roll_period-10-/roll_period-
-roll_range-20-/roll_range-
-pitch_period-30-/pitch_period-
-pitch_range-30-/pitch_range-
-starfield_size-2000-/starfield_size-
-starfield_speed-40-/starfield_speed-
-color_preset-Rainbow-/color_preset-
-start_hue-0-/start_hue-
-hue_change-1-/hue_change-
-/project_specific-
-/project_preferences-
-/account-

I don't think 1. and 2. are connected.

---

_\|/_
U r s

Hi together,

Paul D. Buck just got Problem 2. solved. There was another directory in the Boinc directory named ''account_setup''. When renamed the warning-message is gone.

Thanks Paul

---

_\|/_
U r s

On every startup of Boinc i get the warning message ''Invalid account file : account_setup'' on all of my hosts since i switched a month ago CC4.19 -> CC4.43+.

This may be a long shot, but here goes anyway.
A major problem with the 4.19 clients was the ability to accidentally end up attached to an invalid or blank project.

On one of these hosts that give this error, please open command prompt, go to the BOINC installation folder and run 'boinccmd --get_project_status'.

Do it on a host that has only 1 account_*.xml file, but gives the 'invalid account file" error.
You should get the project status for 1 project and it should match the account_*.xml file. If boinccmd gives you feedback about more than 1 project, let us know.

Paul D. Buck just got Problem 2. solved.

Excellent.

Ignore my previous.

On Problem 1. i just started to download Ethereal (will last about an hour on dial-up). Maybe this tool will be able to give some hints what is happening at ''No schedulers responded'' messages.

---

_\|/_
U r s

On Problem 1.... ''No schedulers responded'' messages.

On the machine that gives this problem, if you type 'ping setiboinc.ssl.berkeley.edu' does it resolve that name to IP 66.28.250.124?