LLRnet servers for NPLB

[gd_barnes]

Originally Posted by henryzz

when will sieving be deep enough for us to begin testing the low ranges for k>2000

June 20th is what we are shooting for.

I'll be rolling out a public sieving drive later today although won't have the file ready to send to everyone until Sunday. That said, I may send out the file with factors we've found so far removed so that people can test how fast it will be on their machines so that we can hone in on that June 20th date.


Gary

[MyDogBuster]

Seems that the GB ports didn't roll over last night. Email notification worked.

[gd_barnes]

Max, your fix for one problem created another. Can you check into the rollover problem please? Thanks.

[mdettweiler]

Okay, I think I know what went wrong. The results kept being piled up in the intermediate files that are normally cleaned out at the end of the day, so the DB pulled them in normally. However, the daily copy-off script didn't run, possibly due to the tweak I made in crotab that was supposed to fix our 1-minute-gap problem. Looks like crontab doesn't like semicolons in command strings.

To work around this I placed the three commands I wanted to run in direct succession (namely, the status page script, the status page script for my private servers, and the copy-off script) in a "daily-run.sh" shell script file and set crontab to run that at 00:01 every day. I verified that the daily-run.sh script does work (and in the process triggered last night's rollover, albeit somewhat belated).

Hopefully this will do the trick--at any rate, we'll find out tomorrow at 00:01 Central time.

[gd_barnes]

OK, cool.

We still have the problem with the result and prime that did not get in the database.

Max, can you follow up with AMDave on that through a PM.


Thanks,
Gary

[mdettweiler]

Originally Posted by gd_barnes

Max, can you follow up with AMDave on that through a PM.

OK, will do.

[MyDogBuster]

GB ports didn't roll over again.

[mdettweiler]

Originally Posted by MyDogBuster

GB ports didn't roll over again.

I've rolled it over manually and put a fix in place that *should* do the trick. Of course, I won't know whether it works until tomorrow's rollover time...

[mdettweiler]

Before anyone else says it--yes, I know, the GB servers didn't roll over for yet another day. I rolled it over manually, but I haven't the faintest idea why cron can't do it as well as I can. I'm running it the exact same way cron is, but for some reason cron seems to be allergic to the copy-off script.

Update: hmm, I just noticed something strange. It looks like cron isn't running *anything* I've put in it--whether it be the copy-off script or even the status page script. It's almost like the whole crontab is shot or something. Any other Linux geeks got an idea what's going on?

Tomorrow I'll try removing the crontab and entering the information afresh. Possibly that will do the trick.

Oh, and because I just know Gary's going to be wondering about this: no, this won't affect the results or miss anything at all. The worst that can happen is for the files to be copied off a few hours late.

[IronBits]

service crond restart

[mdettweiler]

Originally Posted by IronBits

service crond restart

Ah, I hadn't thought of that. I restarted the cron service; I guess we'll find out in a few minutes whether it worked when the status page is due to refresh next.