in the root crontab, about the follozing entries (currently commented out by me): #CZ30 0 * * * /root/check.tables #CZ22 23 * * * du -cs /* > du_slash_`date -I` #CZ23 22 * * * du -cs /home/aras/backup/* > du_home_aras_backup_`date -I` #CZ27 22 * * * du -cs /home/aras/acis/* > du_home_aras_acis_`date -I` #CZ33 22 * * * du -cs /home/* > du_slash_home_`date -I` #CZ36 01 * * * du -cs /var/lib/mysql/* > du_var_lib_mysql_`date -I` #CZ39 01 * * * du -cs /var/* > du_slash_var_`date -I` Are they any good? The machine crashed twice after du -cs /* > du_slash_`date -I`, as far I can see, and I just tried du -cs /home/* > du_slash_home_`date -I` which took several minutes with full load, and the machine was not responsive for a while after that. This worries me. Christian Zimmermann FIGUGEGL! Department of Economics University of Connecticut 341 Mansfield Road, Unit 1063 Storrs, CT 06269-1063 http://ideas.repec.org/zimm/ christian.zimmermann@uconn.edu http://ideas.repec.org/e/pzi1.html
speaking of du -cs ...: of course, it traverses the filesystem -- that's heavy. We tried to use it to analyze where the diskspace was going. They shouldn't have cause the system fault by itself, but who knows? may be it have consumed all the memory while going through the filesystem. (But why would it result in ext3 errors -- i can't imagine.) -ivan On 30 Jan 2008, at 17:24, Christian Zimmermann wrote:
in the root crontab, about the follozing entries (currently commented out by me):
#CZ30 0 * * * /root/check.tables #CZ22 23 * * * du -cs /* > du_slash_`date -I` #CZ23 22 * * * du -cs /home/aras/backup/* > du_home_aras_backup_`date -I` #CZ27 22 * * * du -cs /home/aras/acis/* > du_home_aras_acis_`date -I` #CZ33 22 * * * du -cs /home/* > du_slash_home_`date -I` #CZ36 01 * * * du -cs /var/lib/mysql/* > du_var_lib_mysql_`date -I` #CZ39 01 * * * du -cs /var/* > du_slash_var_`date -I`
Are they any good? The machine crashed twice after du -cs /* > du_slash_`date -I`, as far I can see, and I just tried du -cs /home/ * > du_slash_home_`date -I` which took several minutes with full load, and the machine was not responsive for a while after that. This worries me.
Christian Zimmermann FIGUGEGL! Department of Economics University of Connecticut 341 Mansfield Road, Unit 1063 Storrs, CT 06269-1063 http://ideas.repec.org/zimm/ christian.zimmermann@uconn.edu http://ideas.repec.org/e/pzi1.html
_______________________________________________ RAS-run mailing list RAS-run@lists.openlib.org http://lists.openlib.org/cgi-bin/mailman/listinfo/ras-run
-ivan
On Wed, 30 Jan 2008, Ivan Kurmanov wrote:
speaking of du -cs ...: of course, it traverses the filesystem -- that's heavy. We tried to use it to analyze where the diskspace was going.
They shouldn't have cause the system fault by itself, but who knows? may be it have consumed all the memory while going through the filesystem. (But why would it result in ext3 errors -- i can't imagine.)
It may be a problem when other things are running at the same time. I notice that several of the daily tasks take a long time to run at full (or more than full) load. If they start overlapping, this can become a serious issue. Therefore, I modified crontab for the aras account like I do it for IDEAS: have it call one daily script that does each task in succession. I find the previous organization where each task is called separately at different, set times very dangerous. I am not putting the 'du' instructions back online. However, I think we can put the service back online, it has been off too long now, and we want to make sure it works live before the usual beginning-of-the-month load. So, Kit, work your DNS magic!
-ivan
On 30 Jan 2008, at 17:24, Christian Zimmermann wrote:
in the root crontab, about the follozing entries (currently commented out by me):
#CZ30 0 * * * /root/check.tables #CZ22 23 * * * du -cs /* > du_slash_`date -I` #CZ23 22 * * * du -cs /home/aras/backup/* > du_home_aras_backup_`date -I` #CZ27 22 * * * du -cs /home/aras/acis/* > du_home_aras_acis_`date -I` #CZ33 22 * * * du -cs /home/* > du_slash_home_`date -I` #CZ36 01 * * * du -cs /var/lib/mysql/* > du_var_lib_mysql_`date -I` #CZ39 01 * * * du -cs /var/* > du_slash_var_`date -I`
Are they any good? The machine crashed twice after du -cs /* > du_slash_`date -I`, as far I can see, and I just tried du -cs /home/* > du_slash_home_`date -I` which took several minutes with full load, and the machine was not responsive for a while after that. This worries me.
Christian Zimmermann FIGUGEGL! Department of Economics University of Connecticut 341 Mansfield Road, Unit 1063 Storrs, CT 06269-1063 http://ideas.repec.org/zimm/ christian.zimmermann@uconn.edu http://ideas.repec.org/e/pzi1.html
_______________________________________________ RAS-run mailing list RAS-run@lists.openlib.org http://lists.openlib.org/cgi-bin/mailman/listinfo/ras-run
-ivan
Christian Zimmermann writes
I am not putting the 'du' instructions back online. However, I think we can put the service back online, it has been off too long now, and we want to make sure it works live before the usual beginning-of-the-month load. So, Kit, work your DNS magic!
Did we not agree that you will put the disk that you got from Bob into the machine first, before we restore service? Then the need to check the disk space is gone, for a while. Cheers, Thomas Krichel http://openlib.org/home/krichel RePEc:per:1965-06-05:thomas_krichel skype: thomaskrichel
On Thu, 31 Jan 2008, Thomas Krichel wrote:
Christian Zimmermann writes
I am not putting the 'du' instructions back online. However, I think we can put the service back online, it has been off too long now, and we want to make sure it works live before the usual beginning-of-the-month load. So, Kit, work your DNS magic!
Did we not agree that you will put the disk that you got from Bob into the machine first, before we restore service? Then the need to check the disk space is gone, for a while.
Believe me, I am working on it, but I have currently double teaching duty, heading a recruitment campaign, configuring a new machine, and trying to meet conference submission deadlines. And catching up on various other things because I was away three weeks.
Cheers,
Thomas Krichel http://openlib.org/home/krichel RePEc:per:1965-06-05:thomas_krichel skype: thomaskrichel
participants (3)
-
Christian Zimmermann -
Ivan Kurmanov -
Thomas Krichel