similar to: load testing and tuning a 4GB RAM server

Hi! I've got FreeBSD 8.3-STABLE/i386 server that can be reliably panicked using just 'squid -k rotatelog' command. It seems the system suffers from the problem described here: http://cxsecurity.com/issue/WLB-2010090156 I could not find any FreeBSD Security Advisory containing a fix. My server has 4G physical RAM (about 3.2G available) and runs squid (about 110M VSS) with 500

I first met this problem when our (60 students) internetgateway refused to boot its new kernel, it was a 4.7-RELEASE. Then i loaded the old kernel and went home to check if my 4.8-STABLE does likewise. And the answer was yes! Both kernels were GENERIC + these options taken from the dummynet man pages: options DUMMYNET options NMBCLUSTERS options HZ When i boot the machine

Hello, I have the following entries in /boot/loader.conf: kern.ipc.shm_use_phys="1" kern.ipc.semmns="500" kern.ipc.semmni="40" kern.ipc.semmap="500" which are set correctly. Unfortunately, the following two entries kern.ipc.shmmax="512000000" kern.ipc.shmall="65526" do not change the corresponding values according to

Andre, Are you going to have time to MFC things from -current for auto-tuning -stable before 9.2? I fear (maybe unnecessarily?) that we are about to ship yet another release that can't do basic 10gigE when sufficient memory exists. If you don't have time, then let me know and I'll see what I can do. -- Alfred Perlstein VP Software Engineering, iXsystems

I have an Asus A8V-MX motherboard with an AMD Athlong 64 X2 3800+ CPU and I'm trying to run powerd to keep it cooler/quieter/greener. I'm running -STABLE (6.1-RC) cvsup'ed a couple of days ago, with a kernel config that consists of the SMP sample plus an atapicam device. I'm loading the cpufreq.ko module in /boot/loader.conf. I've attached my dmesg output and sysctl -a

Hello, I have a FreeBSD 7.0-STABLE amd64 box which gives this message with apache 2.2 very often. Previously the contents of the box was on 6.3-STABLE x86 and I had no such problems. This started right away when we moved to 7, 64bit. FreeBSD web.XXXXX.com 7.0-STABLE FreeBSD 7.0-STABLE #0: Tue Apr 22 02:13:30 UTC 2008 yurtesen@web.XXXXX.com:/usr/obj/usr/src/sys/WEB amd64 Approaching the

Hi all. I prefer to build all ports in /tmp, which is actually a tmpfs filesystem. I have 8G partition. But when I tried to compile Apache OpenOffice system starts shooting processes when I was near 6G of swap. Oct 11 18:12:32 ar1l0u kernel: swap zone exhausted, increase kern.maxswzone Oct 11 18:12:32 ar1l0u kernel: pid 1334 (java), uid 818, was killed: out of swap space Oct 11 18:12:32

Hi all. I updated my old i386 machine and now it says: Oct 8 16:57:12 limbo kernel: warning: total configured swap (2100821 pages) exceeds maximum recommended amount (2041984 pages). Oct 8 16:57:12 limbo kernel: warning: increase kern.maxswzone or reduce amount of swap. I though someone tried to remove that warning with kern.maxswzone tunable altogether... What should I do to fix that? #

Hello, I was wondering if it was possible to limit user access on /proc without having to use securelevels. For some reason chmod 751 /proc (or 750) does nothing. Is this possible on FreeBSD 4.9 ? Can't find anything about it in the manual pages. Just want to prevent lusers from running: for file in /proc/*/cmdline; do cat $file; echo; done Greetz, Jimmy Scott

Hello List, I know FreeBSD 4.x is old..., but we are using on a production system with postgres and apache. The above message is appearing periodically. I googled for the message but found no recommendation for adjusting it. Any suggestions. Thanks, Steve -- "They that give up essential liberty to obtain temporary safety, deserve neither liberty nor safety." (Ben Franklin)

After about a day and a half or so of uptime, I'm getting the aforementioned panic on the server ... better then having it hang solid, but right now I'm not sure if this is replacing it, or just one being triggered earlier then the other ... First scan through Google, I came across some posts talking about NMBCLUSTERS ... since its at the same settings as my other server (the default)

Hi All, Mysql service is going down continously in my system due to lack of memory space. I checked the messages log and found the following error message. All mbuf clusters exhausted, please see tuning(7). I have no idea about mbuf cluster. Can anyone please help me to fix the issue. I hope the information below will help you. Following is the output of top. last pid: 84718; load

I have a build process (which worked at release 7.3) that makes a bootable ISO using a ramdisk image as the boot volume. At release 8 it panics right after reporting the real memory size with: kernel trap 12 with interrupts disabled Fatal trap 12: page fault while in kernel mode ... [thread pid 0 tid 0 ] Stopped at pmap_enter+0x19a: moveq (%rcx),%r14 (I have the text frozen on

If you're one of the people who has cvsup'd to 4.8-stable since August 8th and you've since begun to experience panics on a previously stable system, please apply the attached patch and see if your previous stability has been restored. Please tell me your results. Thanks, Mike "Silby" Silbersack -------------- next part -------------- diff -u -r

Hello, I still cannot get dovecot running with more then 1000 processes, but hard limit is 8192 per user in box. I tried everything, including modifying startup script of dovecot to set ulimit -u 8192. Could it be some dovecot bug or dovecot<>freebsd bug? I also tried to set client_limit=2 in imap service to spawn more imap clients in one process, but still I am over 1000 processes with

Hi all. I'm trying to limit the number of participants in a conference room with the realtime option "maxusers", but it doesn't work. Someone have this option working properly? -- thiagoc "O povo n?o deveria temer o governo. O governo ? quem deveria temer o povo." V de Vingan?a

I have some questions about what needs tuned on a high traffic syslog box. I seem to be dropping quite a few syslog packets. This is a syslog server for a high usage Firewall btw. Nic is a Compaq tl0 4.8-P13 netstat -s -p udp | grep buf 19,762,079 dropped due to full socket buffers uptime 5:28PM up 7 days, 18:30, 2 users, load averages: 0.21, 0.23, 0.23 I though maybe syslogd was the

Hi - I found the "maxusers" defined in meetme.c, but I'm not sure how this value is set. Does anybody know if one can limit the number of users permitted in a meetme conference? I know there's MeetmeCount(), but I'd rather avoid the dialplan logic and just set maxusers instead. Thanks, Noah

> This is fixed in CVS. You might consider upgrading to LLVM CVS, > especially if you are interested in looking at performance issues: several > important performance related patches have gone in since LLVM 1.2. OK, now i've really upgraded to CVS, but it looks like state of sources is "not compilable" :( indeed: *************************** Linking llc release executable

Box with fresh RELENG_7 panic under heavy network load (more than 50k connections). This panics seems to be senfile(2) related, because when sendfile disabled in nginx, I can't reproduce the problem. Backtrace in all cases like this: # kgdb kernel /spool/crash/vmcore.1 GNU gdb 6.1.1 [FreeBSD] Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General