Computer Lockup's
Message boards :
NFS Discussion :
Computer Lockup's
Message board moderation
Author | Message |
---|---|
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
I've had 3 Box's since last night Lock up because of the NFS Wu using upwards of 1GB of Memory. I observed this on all 3 Box's, the Wu uses no CPU & the Memory Requirement just Spikes or keeps climbing until the Computer locks up. All you can do is Abort the offending Wu when noticed and make sure the next one on the Box doesn't do the same thing as I had to Abort 3 straight Wu's on 1 Box for the abnormal Memory Usage. |
Send message Joined: 7 Sep 09 Posts: 5 Credit: 37,812 RAC: 0 |
Do you know you can select (or deselect in your case) this new application that uses a lot of memory in your preferences? Then only usual work that uses less memory will be given to you. La frite |
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
I didn't know that, Thanks, I'll give that a try because I just found 2 more Box's that had the Wu's running over 10 Hr's ... PS: Problem is you can't get work for the Low Memory one probably because that's what most people want. |
Send message Joined: 2 Oct 09 Posts: 50 Credit: 111,128,218 RAC: 0 |
... Our problem is that (1) there are very few Cunningham numbers for which the low memory siever is the correct choice; and (2) projects for numbers that size would finish very quickly, just 2-3 days, which is already difficult to manage for the medium-sized siever (as illustrated by the 59-digit ECM factor in the most recent reservations). I could be wrong, but I don't believe that any of our numbers have used the small-memory siever. I've recently been sieving a bunch of numbers with difficulty 240.0-249.99 on our x86-64 clusters (distributed under condor) --- they also use the medium-sized siever. The NFS@Home numbers all have difficulty 250.0-259.99, still in the range where the medium-sized siever is best. I did one number with the large siever which had difficulty 269, but I'm not sure where the crossover is (I'm not sure that we know). I've only done one number with the low-memory siever, which had difficulty in the 220's. One of our friends is expert at finding interest in those small-sized numbers (as well as providing the assembly code settings used for the large memory siever, which is new). I'll check to see whether there's one or two that we could convince Greg would be worth the over-head. Sorry to hear that the new tasks crashed your machine. We had some discussions with the leader of yoyo@home, who was assuring us that BOINC would be able to track larger memory jobs away from machines on which they'd cause a problem. Sounds like we weren't careful enough. Regards, bdodson |
Send message Joined: 26 Jun 08 Posts: 645 Credit: 475,089,748 RAC: 212,324 |
The lowest memory one, lasieved, is not being currently used and hasn't been for most of the project. lasievee is the one that the project has been using for over a month now, so you should leave this one checked. lasievef is causing the problems in your case. BOINC should not download or start workunits if there is insufficient memory for them, but for some unknown reason it is in your case. Deselect lasievef and all will go back to normal. Greg |
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
All my Computers have at least 2GB of Memory & is why BOINC will Download the lasievef Wu's thinking there is enough Memory to run them. Problem is when your running 4 @ 1 time then there isn't enough Memory & why the Lockup's occur. Even if only 1 lasievef were to run & 3 of the 500MB Type too your still short of memory. The lasievee ones seem to use only 357-358MB of Memory & is why I can get away with running 4 of them with only 2GB of Memory on some Box's. Before I went to bed last night I Aborted all the lasievef one's (Sorry) and my Box's ran okay overnight with none Locked up this morning so that's the route for me I guess. |
Send message Joined: 26 Jun 08 Posts: 645 Credit: 475,089,748 RAC: 212,324 |
BOINC shouldn't try to start workunits where the total memory needed is over 2GB, but it apparently did. Anyway, I'm glad it's fixed for you. And feel free to abort workunits as needed. They are automatically reissued. |
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
But does BOINC Know which Wu's will use 1GB & which will only use 357MB, it may get confused with different requirements for Wu's from the same Project ??? |
Send message Joined: 26 Jun 08 Posts: 645 Credit: 475,089,748 RAC: 212,324 |
Yes it does. One of the properties of each workunit is how much memory it will require. |
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
Okay, but the 1GB Wu's could still start up because you have brief periods where the NFS Wu are using hardly any memory right after they start up. It takes a few Minutes for the lasievee Wu's to jump up to 357-358, if a 1GB lasievef Wu were to start up during that brief period then BOINC probably thinks there's enough Memory for it to run when there isn't once all 4 Wu's get up to full Memory Usage .... |
Send message Joined: 9 Oct 09 Posts: 30 Credit: 26,809,482 RAC: 0 |
Just had 8 of the 1GB lasievef Wu's running on my i7 with 6GB of Memory but it locked up, I didn't Abort them off that Box last night just wanting to see if it would lock up if enough of the 1GB Wu's started running. I Aborted the 1GB Wu's now though after seeing it would lock up ... |