virtual memory management
Dear all, Is there a FBSD command to manage virtual memory? I think my swap size is now a bit too much used: last pid: 19824; load averages: 0.06, 0.05, 0.02 up 50+10:00:17 08:54:00 230 processes: 1 running, 227 sleeping, 2 zombie CPU states: 0.0% user, 0.0% nice, 0.4% system, 0.8% interrupt, 98.8% idle Mem: 232M Active, 27M Inact, 91M Wired, 212K Cache, 60M Buf, 142M Free Swap: 512M Total, 482M Used, 29M Free, 94% Inuse The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. Is there any way to handle swap size usage other than restarting the box? Thank you very much in advance! -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
On Sat, Jan 20, 2007 at 08:57:27AM +0100, Zbigniew Szalbot wrote: Dear all, Is there a FBSD command to manage virtual memory? I think my swap size is now a bit too much used: last pid: 19824; load averages: 0.06, 0.05, 0.02 up 50+10:00:17 08:54:00 230 processes: 1 running, 227 sleeping, 2 zombie CPU states: 0.0% user, 0.0% nice, 0.4% system, 0.8% interrupt, 98.8% idle Mem: 232M Active, 27M Inact, 91M Wired, 212K Cache, 60M Buf, 142M Free Swap: 512M Total, 482M Used, 29M Free, 94% Inuse The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. That should not cause such a thing (wget does not try to fit the whole file in memory, so it won't be pushing stuff out to swap). Look at the process sizes in top to see what is using the swap space - something(s) that is still running is using that 482MB. Probably you have one or more processes that are using a large amount of virtual memory, which is too big to fit in RAM. That's the situation you need to address. Is there any way to handle swap size usage other than restarting the box? kill(1). Kris pgpuL9xkWePKe.pgp Description: PGP signature
Re: virtual memory management
Hello again, The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. That should not cause such a thing (wget does not try to fit the whole file in memory, so it won't be pushing stuff out to swap). Look at the process sizes in top to see what is using the swap space - something(s) that is still running is using that 482MB. I do not see such a process: PID USERNAME THR PRI NICE SIZERES STATETIME WCPU COMMAND 21442 root 2 200 224M 26128K kserel 0:06 0.00% java 896 mysql 16 200 70756K 14764K kserel 255:35 0.00% mysqld 98693 bind 1 960 32812K 32172K select 0:22 0.00% dnscache 5035 www 1 40 28372K 15660K accept 5:05 1.86% httpd 5026 www 1 40 28240K 15104K accept 4:46 0.00% httpd 5065 www 1 40 28128K 15196K accept 4:29 0.00% httpd 5030 www 1 40 27892K 15144K accept 4:21 0.00% httpd 5126 www 1 40 27784K 14864K accept 4:20 0.00% httpd 5029 www 1 40 27760K 14644K accept 4:22 0.00% httpd 5027 www 1 40 27740K 15140K accept 4:30 0.00% httpd 5028 www 1 40 27516K 14812K accept 4:03 0.00% httpd 95977 www 1 40 27216K 14532K accept 2:21 0.00% httpd 703 root 1 960 16412K 2296K select 4:35 0.00% httpd 91014 mailman 1 80 11492K 1600K nanslp 6:00 0.00% python 91012 mailman 1 80 11024K 1560K nanslp 5:32 0.00% python 91010 mailman 1 80 11008K 1568K nanslp 5:23 0.00% python 91009 mailman 1 80 11008K 1552K nanslp 5:20 0.00% python I have just restarted the java program as it used to hold 224M but it did not help (and it is quite stable and never given me any problem). After using Squirrel Mail for some time swap use went down to 71% Mem: 258M Active, 49M Inact, 108M Wired, 16M Cache, 60M Buf, 61M Free Swap: 512M Total, 440M Used, 71M Free, 86% Inuse Thank you very much for your suggestion Kris! -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
On Sat, Jan 20, 2007 at 09:13:48AM +0100, Zbigniew Szalbot wrote: Hello again, The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. That should not cause such a thing (wget does not try to fit the whole file in memory, so it won't be pushing stuff out to swap). Look at the process sizes in top to see what is using the swap space - something(s) that is still running is using that 482MB. I do not see such a process: PID USERNAME THR PRI NICE SIZERES STATETIME WCPU COMMAND 21442 root 2 200 224M 26128K kserel 0:06 0.00% java 896 mysql 16 200 70756K 14764K kserel 255:35 0.00% mysqld 98693 bind 1 960 32812K 32172K select 0:22 0.00% dnscache 5035 www 1 40 28372K 15660K accept 5:05 1.86% httpd 5026 www 1 40 28240K 15104K accept 4:46 0.00% httpd 5065 www 1 40 28128K 15196K accept 4:29 0.00% httpd 5030 www 1 40 27892K 15144K accept 4:21 0.00% httpd 5126 www 1 40 27784K 14864K accept 4:20 0.00% httpd 5029 www 1 40 27760K 14644K accept 4:22 0.00% httpd 5027 www 1 40 27740K 15140K accept 4:30 0.00% httpd 5028 www 1 40 27516K 14812K accept 4:03 0.00% httpd 95977 www 1 40 27216K 14532K accept 2:21 0.00% httpd 703 root 1 960 16412K 2296K select 4:35 0.00% httpd 91014 mailman 1 80 11492K 1600K nanslp 6:00 0.00% python 91012 mailman 1 80 11024K 1560K nanslp 5:32 0.00% python 91010 mailman 1 80 11008K 1568K nanslp 5:23 0.00% python 91009 mailman 1 80 11008K 1552K nanslp 5:20 0.00% python I see lots of them; every one in that list is contributinig. If you add up all those process sizes you'll see where the space is going. Basically you are just overloading your system by trying to run too much at once. Reduce the load or add more RAM. Kris pgpA2qhyW02hO.pgp Description: PGP signature
Re: virtual memory management
On Sat, Jan 20, 2007 at 03:51:38AM -0500, Kris Kennaway wrote: On Sat, Jan 20, 2007 at 09:13:48AM +0100, Zbigniew Szalbot wrote: Hello again, The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. That should not cause such a thing (wget does not try to fit the whole file in memory, so it won't be pushing stuff out to swap). Look at the process sizes in top to see what is using the swap space - something(s) that is still running is using that 482MB. I do not see such a process: PID USERNAME THR PRI NICE SIZERES STATETIME WCPU COMMAND 21442 root 2 200 224M 26128K kserel 0:06 0.00% java 896 mysql 16 200 70756K 14764K kserel 255:35 0.00% mysqld 98693 bind 1 960 32812K 32172K select 0:22 0.00% dnscache 5035 www 1 40 28372K 15660K accept 5:05 1.86% httpd 5026 www 1 40 28240K 15104K accept 4:46 0.00% httpd 5065 www 1 40 28128K 15196K accept 4:29 0.00% httpd 5030 www 1 40 27892K 15144K accept 4:21 0.00% httpd 5126 www 1 40 27784K 14864K accept 4:20 0.00% httpd 5029 www 1 40 27760K 14644K accept 4:22 0.00% httpd 5027 www 1 40 27740K 15140K accept 4:30 0.00% httpd 5028 www 1 40 27516K 14812K accept 4:03 0.00% httpd 95977 www 1 40 27216K 14532K accept 2:21 0.00% httpd 703 root 1 960 16412K 2296K select 4:35 0.00% httpd 91014 mailman 1 80 11492K 1600K nanslp 6:00 0.00% python 91012 mailman 1 80 11024K 1560K nanslp 5:32 0.00% python 91010 mailman 1 80 11008K 1568K nanslp 5:23 0.00% python 91009 mailman 1 80 11008K 1552K nanslp 5:20 0.00% python I see lots of them; every one in that list is contributinig. If you add up all those process sizes you'll see where the space is going. By which I mean the difference between size and res, which indicates the amount of process memory allocated but not currently resident in RAM. This isn't a foolproof method (see e.g. the FAQ entry on rpc.statd), but it's true in your case. Basically you are just overloading your system by trying to run too much at once. Reduce the load or add more RAM. Kris pgpTxEofIrvtn.pgp Description: PGP signature
Re: virtual memory management
Dear Kris and all, I see lots of them; every one in that list is contributinig. If you add up all those process sizes you'll see where the space is going. By which I mean the difference between size and res, which indicates the amount of process memory allocated but not currently resident in RAM. This isn't a foolproof method (see e.g. the FAQ entry on rpc.statd), but it's true in your case. Basically you are just overloading your system by trying to run too much at once. Reduce the load or add more RAM. The problem is I cannot add more RAM (too old machine to do that) but I know what to do to decrease the load a bit. So thanks for the pointer! I appreciate it! Warm regards, -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
Zbigniew Szalbot wrote: The problem is I cannot add more RAM (too old machine to do that) but I know what to do to decrease the load a bit. So thanks for the pointer! I appreciate it! You might also want to stop using mod_php in apache and convert to fastcgi setup - this way you'll get all Apache processes to use a much more reasonable amount, like ~~5MB - 8MB and a small number of php-cgi processes that use ~~20MB or more, saving you memory in the end. signature.asc Description: OpenPGP digital signature
Re: virtual memory management
hello, The problem is I cannot add more RAM (too old machine to do that) but I know what to do to decrease the load a bit. So thanks for the pointer! I appreciate it! You might also want to stop using mod_php in apache and convert to fastcgi setup - this way you'll get all Apache processes to use a much more reasonable amount, like ~~5MB - 8MB and a small number of php-cgi processes that use ~~20MB or more, saving you memory in the end. Does this mean recompiling Apache? Or is it a question of httpd.conf? From what I understand it probably involves recompilation? Thank you! -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
On Saturday 20 January 2007 08:57, Zbigniew Szalbot wrote: Is there any way to handle swap size usage other than restarting the box? Yes, you can add swap while the system is running with swapon(8). If you don't have an empty partition available you could create one with mdconfig(8). -Pieter ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
Zbigniew Szalbot wrote: hello, The problem is I cannot add more RAM (too old machine to do that) but I know what to do to decrease the load a bit. So thanks for the pointer! I appreciate it! You might also want to stop using mod_php in apache and convert to fastcgi setup - this way you'll get all Apache processes to use a much more reasonable amount, like ~~5MB - 8MB and a small number of php-cgi processes that use ~~20MB or more, saving you memory in the end. Does this mean recompiling Apache? Or is it a question of httpd.conf? From what I understand it probably involves recompilation? At most it would require recompilation of PHP (the main port, not the extensions) and installing mod_fcgid. If you enabled CGI and FastCGI during PHP build, you only need mod_fcgid. See http://fastcgi.coremail.cn/configuration.htm#PHP for documentation. (don't forget to remove mod_php from httpd.conf). Note that using FastCGI is different in some important aspects from mod_php, so read up on it if you haven't used it before. signature.asc Description: OpenPGP digital signature
Re: virtual memory management
In the last episode (Jan 20), Zbigniew Szalbot said: I see lots of them; every one in that list is contributinig. If you add up all those process sizes you'll see where the space is going. By which I mean the difference between size and res, which indicates the amount of process memory allocated but not currently resident in RAM. This isn't a foolproof method (see e.g. the FAQ entry on rpc.statd), but it's true in your case. Basically you are just overloading your system by trying to run too much at once. Reduce the load or add more RAM. The problem is I cannot add more RAM (too old machine to do that) but I know what to do to decrease the load a bit. So thanks for the pointer! I appreciate it! Also remember that swap usage itself is not a bad thing; it just means the system has moved some unused process data to disk. What /is/ bad is when the system is so low on RAM that is it constantly shuffling data to and from swap just to keep running. This is called thrashing, and you can track it by watching the ##Kb In, ##Kb Out values on the Swap: line in top, and the pi and po columns in vmstat output. As long as you don't see constant swapping activity, you're okay. -- Dan Nelson [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
On Sat, Jan 20, 2007 at 08:57:27AM +0100, Zbigniew Szalbot wrote: Dear all, Is there a FBSD command to manage virtual memory? I think my swap size is now a bit too much used: last pid: 19824; load averages: 0.06, 0.05, 0.02 up 50+10:00:17 08:54:00 230 processes: 1 running, 227 sleeping, 2 zombie CPU states: 0.0% user, 0.0% nice, 0.4% system, 0.8% interrupt, 98.8% idle Mem: 232M Active, 27M Inact, 91M Wired, 212K Cache, 60M Buf, 142M Free Swap: 512M Total, 482M Used, 29M Free, 94% Inuse The swap size usage grow so big probably because I started wget to download an iso image and then WinSCP to grab it from the FBSD machine to my laptop. When I started wget, the swap usage was around 19% and had been like that for many days. Is there any way to handle swap size usage other than restarting the box? Don't forget that the system also pages to swap space and it takes the attitude of parking as much as possible out there in case it comes in to demand again. Ten if it really needs the space for something, it invalidates the oldest stuff and uses that space. So, you should really expect that your swap space should be nearly maxed all the time if things are working well. Someone else can give you more accurate detailed information if you want it. jerry Thank you very much in advance! -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
Dear all, | Also remember that swap usage itself is not a bad thing; it just means Problem solved. I should have thought about that earlier. Yesterday I was playing with HotSaNIC software to use it on this box. In the end I decided I didn't like it and I didn't really need it so I removed it from the system stopping rrdtool first. Only today did I notice that I have an unusually high (for my setup) number of sleeping processes. ps ax showed me that I had about 150 perl processes that were started by HotSaNIC but never really stopped. I killed them (what a joy :) and voila: last pid: 62059; load averages: 0.04, 0.12, 0.22 up 51+00:27:42 23:21:25 81 processes: 1 running, 78 sleeping, 2 zombie CPU states: 2.7% user, 0.0% nice, 0.8% system, 0.8% interrupt, 95.7% idle Mem: 161M Active, 4348K Inact, 111M Wired, 128K Cache, 60M Buf, 216M Free Swap: 512M Total, 81M Used, 431M Free, 15% Inuse Thank you all for your support! My experience with FBSD is pretty much 51 days old :). But I read most of the posts hoping to learn new things. The great adventure now is to upgrade to 6.2 from 6.1. Never been there before :) Warm regards, -- Zbigniew Szalbot ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
Don't forget that the system also pages to swap space and it takes the attitude of parking as much as possible out there in case it comes in to demand again. Ten if it really needs the space for something, it invalidates the oldest stuff and uses that space. So, you should really expect that your swap space should be nearly maxed all the time if things are working well. If this is the case something is really wrong on my system: Swap: 4096M Total, 4096M Free either top doesn't show the precautionary swapping or this is not happening. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: virtual memory management
[LoN]Kamikaze wrote: Don't forget that the system also pages to swap space and it takes the attitude of parking as much as possible out there in case it comes in to demand again. Ten if it really needs the space for something, it invalidates the oldest stuff and uses that space. So, you should really expect that your swap space should be nearly maxed all the time if things are working well. If this is the case something is really wrong on my system: Swap: 4096M Total, 4096M Free either top doesn't show the precautionary swapping or this is not happening. Precautionary swapping does exist, but it's not that often :) What the poster probably meant is that it's possible to have a perfectly working system which looks like it's using a lot of swap if the swapin/swapout rate is low enough. signature.asc Description: OpenPGP digital signature
