Weekend server problems (Oct 1, 2007)

Author	Message
Andy Lee Robinson Send message Joined: 8 Dec 05 Posts: 630 Credit: 59,973,836 RAC: 0	Message 654262 - Posted: 5 Oct 2007, 0:05:49 UTC Last modified: 5 Oct 2007, 0:07:32 UTC Bruno is up, downloading is fine. Uploading is painful with 1 in 100 retries successful. Can this NFS driver actually be examined and cured? Perhaps it's a tcp MTU problem again? Is there anyone around to give it a kick? ID: 654262 ·

1mp0Â£173 Volunteer tester Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0	Message 654316 - Posted: 5 Oct 2007, 2:18:19 UTC - in response to Message 654262. Bruno is up, downloading is fine. Uploading is painful with 1 in 100 retries successful. Can this NFS driver actually be examined and cured? Perhaps it's a tcp MTU problem again? Is there anyone around to give it a kick? From the home page: October 4, 2007 We tried moving the upload store off of a possibly flaky disk array yesterday. This slowed upload service dramatically. The slowdown was unexpected and we are in the process of fixing it.. ID: 654316 ·

Andy Lee Robinson Send message Joined: 8 Dec 05 Posts: 630 Credit: 59,973,836 RAC: 0	Message 654331 - Posted: 5 Oct 2007, 2:44:59 UTC - in response to Message 654316. Ah! :-) Thanks - it never says anything new when I look at it! ID: 654331 ·

Lee Gresham Send message Joined: 12 Aug 03 Posts: 159 Credit: 130,116,228 RAC: 0	Message 654368 - Posted: 5 Oct 2007, 3:32:43 UTC I am also having trouble with uploads and reporting tasks on at least a dozen machines. No way its my hardware on that many. Got to be with one of the Seti machines Delta-V ID: 654368 ·

Andy Lee Robinson Send message Joined: 8 Dec 05 Posts: 630 Credit: 59,973,836 RAC: 0	Message 654377 - Posted: 5 Oct 2007, 3:53:23 UTC - in response to Message 654368. I am also having trouble with uploads and reporting tasks on at least a dozen machines. No way its my hardware on that many. Got to be with one of the Seti machines Everyone is - see home page... It'll clear up.. main thing is that downloads are still working so no noone will run out of work even if they have a small cache to optimise turnover. ID: 654377 ·

Tilo Send message Joined: 2 Oct 07 Posts: 1 Credit: 175,148 RAC: 0	Message 654467 - Posted: 5 Oct 2007, 10:13:11 UTC Hello all together, The Problem seems to be fixed. Right now my BOINC Client is up and downloading. and it was time. In around 3 hours I would have finished all tasks. ID: 654467 ·

KWSN THE Holy Hand Grenade! Volunteer tester Send message Joined: 20 Dec 05 Posts: 3187 Credit: 57,163,290 RAC: 0	Message 654618 - Posted: 5 Oct 2007, 16:28:09 UTC - in response to Message 654251. Just on that note though, I've been getting random computation errors and all sorts of fun things, like SETI reporting work done when I never saw it even starting to work. Dunno if it's corrupted work from the server sending and losing too many packets, though I didn't think so, since I thought BOINC would use TCP/IP to transmit packets. A) Yes, BOINC uses TCP/IP. B) Can't help with the "random computation errors"... C) The "reporting work done" that you never saw the computer do is probably canceled WU's . Say that someone goes past deadline and a replacement is issued (to you); then gets the WU in after deadline, but before your computer gets to that WU in your cache. Your WU is now redundant, and will be canceled. You won't get any credit for it, but then you never crunched on it either... . Hello, from Albany, CA!... ID: 654618 ·

©2025 University of California

SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.