Lost WU's

[chawken]

I have Seti Driver configured to 'Create SetiLog.csv'. Every afternoon at about 4 oclock I import the results from all my machines to a custom Excel sheet. According to the combined logs, SetiLog shows that I should have transmitted 53 WU's in the last 24hours, but Berkeley only shows 47. That is a loss of 6 WU's in the last 24hours. The last time that I lost WU's was when Berkeley changed servers, awhile back, but that was only about 4 in that whole week.

Anyone else losing WU's in the last day or so?

 

[[Oc]acaridans]

Ive probally lost 6 WU's in total..I lost 1 last week...but none since then

 

[TC]

I've lost 8 in the past 48 hours

 

[Morpheus]

hmm... my seti driver has been hanging... I had to "kill" the result.sah file WU #16 to get Driver up & running again...

This is bad news...

Our apologies to those of you using parsers or 3rd party software to process our stats pages. We made some HTML changes to those pages that apparently broke many of your algorithms. On the bright side, we've reverted to the original (pre-8/13) formatting, and our team pages are in the process of being reverted as well.

From the Berekley server..

 

[Morpheus]

Roelof (@ SETI Spy) says:

8/15/2001: User Stats Update Broken (again). Due to some changes to SETI@home's old user stats web page format, SETI Spy's User Stats Update function now returns "unknown" for most of the statistics. I have emailed Berkeley about this, and will decide on what I need to do when I hear from them. They can either change back to page to the old format (I hope, because this would not require any change to SETI Spy), or I need to update SETI Spy to support the new format. Keep watching this space for updates!

Update: The problem mentioned above has been fixed. According to a post by SETI@home's Eric Person on the alt.sci.seti newsgroup: "... I changed back the user_stats formatting to its original state, so SETI Spy and other parsing algorithms using the user_stats cgi should be ok now." Many thanks to the SETI@home team for quickly addressing this problem.

Crunch On!!!

 

[TC]

Off topic here, but any idea where Roelof is from? I had the impression he lived in Texas because of his email address. I figured his name was just a handle of some sort, so one day I asked him what his real name is. Of course it's Roelof - assuming that's a last name? Hope he wasn't offended, but that's an intersting name to me.

 

[chawken]

Roelof (@ SETI Spy) says:



Crunch On!!!

This isn't a SetiSpy problem that I'm having. I use SetiDriver's SetiLog.csv. But because of the discrepancy between my logs and what Berkeley is reporting, I have been watching my machines and checking when they turn in work units. Since my post yesterday, till this morning (7.00am pst) I have lost another 8 units. This is a little upsetting, 14 WU's lost in the last 36hrs.

 

[Morpheus]

I guess i am being a bit too cryptic here.. sorry... I was suggesting by the above that there are changes a happenin' @ the Berekely server, which almost always results in some soft of trouble

Might try backing up your result.sah files and doing a manual upload if that continues??? I know it is a drag, but it might save lost cycles??

Good luck!!!

 

[chawken]

I guess i am being a bit too cryptic here.. sorry... I was suggesting by the above that there are changes a happenin' @ the Berekely server, which almost always results in some soft of trouble

Might try backing up your result.sah files and doing a manual upload if that continues??? I know it is a drag, but it might save lost cycles??

Good luck!!!

Thanks Kris. I'll watch my machines carefully this weekend, and if it continues I'll turn off the Auto Transmit for awhile.