Re: PID 11 using 400% CPU
Hello Dan,
I was having multiple problems with my x86 installation, apart from
the disc spinning continuously. Despite tuning many sysctl parameters,
X clients would not open for non-root users ("max number of clients
reached"); for root, they would open, but after about 30 minutes of
playing xboard (gnuchess), the computer would simply shut down
(without even a proper shut down sequence; worse that DOS BSOD - power
off in a flash); while up, the speed and performance of the system was
abysmal.
I therefore yesterday downloaded the iso for amd64, which is the
architecture of my system. I am going to install it and try it out
over the weekend. Hope things will be better this time around.
Thanks for all your suggestions and advice though.
Regards
Manish Jain
[1]invalid.poin...@gmail.com
On 05-Jul-11 19:50, Dan Nelson wrote:
In the last episode (Jul 05), manish jain said:
On 30 June 2011 10:26, Dan Nelson [2] wrote:
In the last episode (Jun 30), Manish Jain said:
I have a strange problem with my 8.1 box. After booting, the hard
disk goes into a full-speed never-ending spin.
To see what disk I/O is being done, try running "ktrace -dip 0 ; sleep
10 ; ktrace -C", to capture all syscalls done on the entire system (pid
0 plus children) for 10 seconds, then run "kdump -m64 | less" to view
the results. Look for read or write calls.
It looks like ppp is doing a lot of read and write operations, which keeps
the disk spinning. How do I set this right ? Is there something wrong
with my ppp.conf (see below) ?
I bet that if you ran fstat or lsof on the ppp process, all the writes are
actually to your serial device or a tun device, not to disk. ppp is
unlikely to cause much disk I/O. You'll have to filter out the ppp process
and check your kdump output again.
References
1. mailto:invalid.poin...@gmail.com
2. mailto:dnel...@allantgroup.com
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: fubar'ed it good this time...
On Mon, Jul 4, 2011 at 15:30, Dr. A. Haakh wrote: > Kurt Buff schrieb: >> >> On Thu, Jun 30, 2011 at 17:31, Kurt Buff wrote: >> >>> >>> On Tue, Jun 28, 2011 at 06:50, Dr. A. Haakh wrote: >>> Polytropon schrieb: > > On Mon, 27 Jun 2011 06:40:27 -0700, Kurt Buff wrote: > > >> >> Your advice sounds reasonable, but that site seems devoted to zfs >> bootables. >> >> I wonder if an 8.1 livefs iso will do the trick... >> >> > > Check if you can download FreeSBIE somewhere. It's a live system > using the 5.x and 6.x kernel which should be fine. Next to two > GUI modes (light, heavy) it also has a versatile "maintenance mode" > for such operations. I have already successfully used this system > for solving similar situations, for diagnostics, and for data > recovery preparation. > The loader obviously knows how to deal with the filesystem because he loads the failing new kernel. So the easiest solution would be to boot an older kernel if available. I don't know how freebsd-update deals with older kernels, he should still be around. First guess is /boot/kernel.old/kernel. So get the loader-prompt, "unload kernel" and try "load /boot/kernel.old/kernel". Andreas >>> >>> OK - to continue, while I have a few free minutes. >>> >>> I have been able to load the old kernel by going to the loader prompt >>> from the boot menu, and doing >>> unload kernel >>> load /boot/kernel.old/kernel >>> >>> That barked about linproc in fstab, so I edited that out. >>> >>> Then, the next go-round: It complained about mismatches on >>> daemon_saver.ko - a version mismatch, so I've commented that out of >>> /etc/rc.conf. It also complained about linux.ko, so that's been >>> commented out in /etc/rc.conf as well. >>> >>> I'm now able to reboot cleanly with the old kernel. >>> >>> After doing 'freebsd-update install' for the second time, I still >>> can't get 8.2 to boot - same issue, only acd0 is recognized. However, >>> I'm logged in as root under the old kernel, though I haven't start >>> XFCE4, and don't have wireless running. >>> >>> This one is getting to be fun... >>> >>> Kurt >>> >> >> So, I tried booting from the old kernel again, and then did a >> 'freebsd-update rollback', and that worked just fine. I thought I'd >> try again, but first did a 'freebsd-update fetch' and 'freebsd-update >> install' to get the latest 8.1 updates. >> >> That worked just fine, so I did a 'freebsd-update -r 8.2-RELEASE >> fetch' again, then a 'freebsd-update install', which went just fine, >> and after that rebooted as directed to attempt the second >> 'freebsd-update install'. >> >> That's when the same thing happened - i got dumped into the mountroot >> prompt again. And, again, rebooting and escaping to the loader prompt >> allows me to unload the kernel, load /boot/kernel.old/kernel then >> autoboot, and boot up. Same as before. >> >> Any thoughts? >> > > Redo the rollback to 8.1 and install the 8.2-STABLE source-tree. You can > install the 8.1-sources from cd and update them to 8.2-STABLE using csup. > > Put the following lines in /etc/make.conf > SUP_UPDATE= YES > SUP= /usr/bin/csup > SUPHOST= cvsup2.de.freebsd.org > SUPFILE= /usr/share/examples/cvsup/stable-supfile > > Make sure that stable-supfile contains the right tag > *default release=cvs tag=RELENG_8 > > Then goto /usr/src and "make update |tee _Update.log" > > Once the source-tree is up-to-date: > > Have a look at the FreeBSD Handbook: Chapter 8 - Configuring the FreeBSD > Kernel. > > Copy GENERIC to e.g. MYKERNEL, edit MYKERNEL and add some debug-flags. See > /sys/conf/NOTES for additional debug-options. You may as well try GENERIC - > maybe your problem is gone... > Then go to /usr/src and do s.th. like "make buildkernel KERNCONF=MYKERNEL" > and if the kernel built fine install it: "make KERNCONF=YOURKERNEL > installkernel". You can also add KERNCONF=MYKERNEL to /etc/make.conf instead > of adding it to the make command. > Installing the new kernels moves /boot/kernel to /boot/kernel.old and > installs the new one in /boot/kernel. If the new kernel fails again, you can > delete it: "rm -rf /boot/kernel && mv /boot/kernel.old /boot/kernel" thus > putting the previous kernel in the right place. > If the new kernel fails again, then press the Scroll-key and navigate to the > disk-probe usind page-up-key. Write down the messages or take a photo and > post it to this list. > > If the STABLE kernel boots fine you will probably want to remove all the > debugging stuff and rebuild it. > > If you intend to keep the 8.1-kernel move it to e.g /boot/kernel-8.1 so it > will not be deleted, when you install new ones and you can always "load > /boot/kernel-8.1/kernel" from the loader > > Once the new kernel boots fine, cd /usr/src and follow the instructions in > Makefile how to build and install a new kernel an
Re: Tool to show the recent disk space consumers?
On Tue, Jul 5, 2011 at 12:37, Yuri wrote: > Hi, > > I hit this problem periodically when a lot of disk space is gone and it's > hard to tell where did it go. Once it was thunderbird writing huge index > file as a consequence of some bug, on another occasion it was the bug in KDE > writing some huge index somewhere in ~/.kde4. > > Is there a tool slowly indexing the file system and showing where exactly > did the sudden growth of consumed space occur? > > I know about du(1) but I am looking for some program that can detect the > dynamics and pinpoint the offending files. > > Yuri kdirstat might prove useful, if it's run periodically. Kurt ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
New games for you
Hi Elias! Try out these new games! Happy Town, played 63 272 times Let's play! http://en.netlog.com/go/mailurl/-YT04NjQ5MzUyM2MzZDRjYTZlY2FjMTkxMzA1ZDhmNWQ2YSZtPTE5NTgzNTIwNTUmbD0xJmdtPTUxJnU9JTJGcGxheSUyRmdhbWUlMkZoYXBweV90b3du You no longer want to receive these updates by e-mail? Change your settings. http://en.netlog.com/go/mailurl/-YT04NjQ5MzUyM2MzZDRjYTZlY2FjMTkxMzA1ZDhmNWQ2YSZtPTE5NTgzNTIwNTUmbD0yJmdtPTUxJnU9JTJGZ28lMkZzZXR0aW5ncyUyRmVtYWls Your Netlog account: http://www.netlog.com/shashatielias Massive Media NV E. Braunplein 18. B-9000 Gent. Belgium BE0859635972. abuse...@netlog.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Tool to show the recent disk space consumers?
Yuri writes: > I hit this problem periodically when a lot of disk space is gone and > it's hard to tell where did it go. Once it was thunderbird writing > huge index file as a consequence of some bug, on another occasion it > was the bug in KDE writing some huge index somewhere in ~/.kde4. > > Is there a tool slowly indexing the file system and showing where > exactly did the sudden growth of consumed space occur? > > I know about du(1) but I am looking for some program that can detect > the dynamics and pinpoint the offending files. A quick-and-rough approach that I sometimes use is to look at the incremental backups... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Tool to show the recent disk space consumers?
> > I hit this problem periodically when a lot of disk space is gone and it's > hard to tell where did it go. Once it was thunderbird writing huge index > file as a consequence of some bug, on another occasion it was the bug in KDE > writing some huge index somewhere in ~/.kde4. > > Is there a tool slowly indexing the file system and showing where exactly > did the sudden growth of consumed space occur? > > I know about du(1) but I am looking for some program that can detect the > dynamics and pinpoint the offending files. > A lot of the monitoring applications can do this such as zabbix and opennms, but they don't necessarily monitor file size on every file so I don't know how efficient it would be if it's not integrated into kqueue. One of the security applications such as samhain may be able to do a better job of it. -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Tool to show the recent disk space consumers?
Hi, I hit this problem periodically when a lot of disk space is gone and it's hard to tell where did it go. Once it was thunderbird writing huge index file as a consequence of some bug, on another occasion it was the bug in KDE writing some huge index somewhere in ~/.kde4. Is there a tool slowly indexing the file system and showing where exactly did the sudden growth of consumed space occur? I know about du(1) but I am looking for some program that can detect the dynamics and pinpoint the offending files. Yuri ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: PID 11 using 400% CPU
In the last episode (Jul 05), manish jain said: > On 30 June 2011 10:26, Dan Nelson wrote: > > In the last episode (Jun 30), Manish Jain said: > > > > > >I have a strange problem with my 8.1 box. After booting, the hard > > >disk goes into a full-speed never-ending spin. > > > > To see what disk I/O is being done, try running "ktrace -dip 0 ; sleep > > 10 ; ktrace -C", to capture all syscalls done on the entire system (pid > > 0 plus children) for 10 seconds, then run "kdump -m64 | less" to view > > the results. Look for read or write calls. > > It looks like ppp is doing a lot of read and write operations, which keeps > the disk spinning. How do I set this right ? Is there something wrong > with my ppp.conf (see below) ? I bet that if you ran fstat or lsof on the ppp process, all the writes are actually to your serial device or a tun device, not to disk. ppp is unlikely to cause much disk I/O. You'll have to filter out the ppp process and check your kdump output again. -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: ipfw: getsockopt(IP_FW_ADD): Invalid argument
--- On Tue, 7/5/11, Ian Smith wrote: > From: Ian Smith > Subject: Re: ipfw: getsockopt(IP_FW_ADD): Invalid argument > To: "Unga" > Cc: freebsd-questions@freebsd.org > Date: Tuesday, July 5, 2011, 5:14 PM > In freebsd-questions Digest, Vol 370, > Issue 2, Message: 14 > On Mon, 4 Jul 2011 09:11:07 -0700 (PDT) Unga > wrote: > > --- On Mon, 7/4/11, Unga > wrote: > > > > > From: Unga > > > Subject: ipfw: getsockopt(IP_FW_ADD): Invalid > argument > > > To: freebsd-questions@freebsd.org > > > Date: Monday, July 4, 2011, 11:48 AM > > > Hi all > > > > > > Following ipfw rule develop error indicated in > the subject > > > line: > > > ipfw add 100 fwd 127.0.0.1,1234 tcp from any to > any 1234 > > > out via wlan0 > > > > > > What I want to do is forward any packet going to > port 1234 > > > to 127.0.0.1:1234. > > > > > > I have built the kernel with "options > > > IPFIREWALL_FORWARD". > > > > > > What's the error here? Is the rule incorrect? > > > > > > This is FreeBSD 8.1. > > > > > > Many thanks in advance. > > > > > > Unga > > > > > > > Does anybody successfully use the "ipfw fwd"? If so > in which FreeBSD version? > > Not I, but many do. On the face of it the rule looks > correct. Do you > have a TCP service running on localhost:1234 ? Does > wlan0 exist? You > may do better posting to the freebsd-ipfw list, with more > information. > > cheers, Ian Hi Ian Thanks for the reply. I do have a daemon listening to the port 1234 on the local machine. The wlan0 exists and ip number 192.168.1.3 assigned. I'll write the freebsd-ipfw. Thank you. Unga ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: ipfw: getsockopt(IP_FW_ADD): Invalid argument
In freebsd-questions Digest, Vol 370, Issue 2, Message: 14 On Mon, 4 Jul 2011 09:11:07 -0700 (PDT) Unga wrote: > --- On Mon, 7/4/11, Unga wrote: > > > From: Unga > > Subject: ipfw: getsockopt(IP_FW_ADD): Invalid argument > > To: freebsd-questions@freebsd.org > > Date: Monday, July 4, 2011, 11:48 AM > > Hi all > > > > Following ipfw rule develop error indicated in the subject > > line: > > ipfw add 100 fwd 127.0.0.1,1234 tcp from any to any 1234 > > out via wlan0 > > > > What I want to do is forward any packet going to port 1234 > > to 127.0.0.1:1234. > > > > I have built the kernel with "options > > IPFIREWALL_FORWARD". > > > > What's the error here? Is the rule incorrect? > > > > This is FreeBSD 8.1. > > > > Many thanks in advance. > > > > Unga > > > > Does anybody successfully use the "ipfw fwd"? If so in which FreeBSD version? Not I, but many do. On the face of it the rule looks correct. Do you have a TCP service running on localhost:1234 ? Does wlan0 exist? You may do better posting to the freebsd-ipfw list, with more information. cheers, Ian___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: the alternative function for inet_aton
On Mon, Jul 4, 2011 at 8:05 PM, ahmad javadi wrote: > hi > I use " inet_aton" function in the kernel socket programming but i have the > following warrning: > "crypto.c:63: warning: implicit declaration of function 'inet_aton' > crypto.c:63: warning: nested extern declaration of 'inet_aton' > " > this function required that "" be included but this folder did > not exit in the my freebsd. There's probably a typo in your #include: replace the "," with a "." and try again. On my 8.2-STABLE system, /usr/include/arpa/inet.h exists, and defines inet_aton() just fine: % grep 'inet_aton' /usr/include/arpa/inet.h #define inet_aton __inet_aton int inet_aton(const char *, struct in_addr *); > Please help me to find solution or to use alternative function. > thanks -cpghost. -- Cordula's Web. http://www.cordula.ws/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: PID 11 using 400% CPU
On Tue, Jul 5, 2011 at 6:27 AM, manish jain wrote: > Hello Dan, > > It looks like ppp is doing a lot of read and write operations, which keeps > the disk spinning. How do I set this right ? Is there something wrong with > my ppp.conf (see below) ? > > ppp.conf : > > default: > set log Phase Chat LCP IPCP CCP tun command Maybe reduce the amount of logging? But look at the logs before turning logging off: maybe there's something wrong with your setup, perhaps you have a flaky line etc...? -cpghost. -- Cordula's Web. http://www.cordula.ws/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Re: why desktop apps are able to kill my freebsd box?
On Tue, Jul 5, 2011 at 2:42 AM, Timo wrote: >> Why a faulty desktop application run as unprivileged user is able to >> crash my system? >> >> I mean, I know programs have bugs and sometimes they lead to crashes. >> I'm fine with that. But why a crashing program (for example firefox or >> banshee) is able to kill the whole system? >> >> And by 'crash' or 'kill' i mean that for whatever reason the system is >> frozen and doesn't reply to anything but a hard reset. Can you still log in via ssh from another box? Or at least, can you ping the "frozen" box from the outside? Usually, a frozen system isn't really frozen, it's just Xorg which is. And that's often related to buggy graphics drivers. -cpghost. -- Cordula's Web. http://www.cordula.ws/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
