Search results

Any idea how to build MySQL with DTrace support? I've built it with: make -DENABLE_DTRACE install clean or: make -WITH_DTRACE=1 install clean

I did enable it. Also I am looking for a DTrace PHP module (there's one for Solaris). Update: A little bit offtopic but still related to Apache. I've managed to use DTrace with PHP. As of php5-5.4.3 (which is in Ports right now) there is support for DTrace. After you've compiled and installed...

I am looking for a tutorial regarding this topic. I've found this but don't know if it works ok on FreeBSD: http://prefetch.net/projects/apache_modtrace/index.html

I was having the same problem (slow scrub) and after using vfs.zfs.vdev.max_pending="4" the speed increased from 3MB/s to 400MB/s. Update: My mistake, this option did not solved anything. I still have the same slow speed and 400MB/s was in fact for resilvering and not for scrub.

@phoenix: The process works the way you've describe it. And yes, split works on FreeBSD. Next example shows the process: zpool create tank mirror ada1 ada2 zpool attach tank ada2 ada3 zpool scrub tank # this is not really necessary, resilver will be done anyway...

Use OpenDNS DNS's Or set up a DNS server on a machine in a datacenter.

Try mysqltuner Try mysqltuner app from Ports: # cd /usr/ports/databases/mysqltuner # make install clean ; rehash # mysqltuner

My short advice on cleaning a server did not include how to clean dead fish bodies or dead cockroach bodies (which, no joke, I did encounter when cleaning home servers).

@Beeblebrox: Your idea is bad practice: you must use a more powerful jet. Like the one used by firemen.

We must bring the beer!

1. Use a vacuum cleaner that has two holes, one that suck and one that evacuates air. Put the pipe of the vacuum cleaner to the hole that evacuates air. This is the best way to clean it. Then, clear the remaining dust with a brush. 2. Clear the dust from inside the cooler with medicinal...

Read dinoex@'s solution. That worked for me.

The ps line might be longer, the one I've posted is just what I was able to grep(1).

Well, the ps code I've posted in my first post on this thread does that every morning at 3 AM. I wonder which variable from /etc/defaults/periodic.conf file (if any) should I copy to /etc/periodic.conf and modify accordingly to skip this process for /mnt dir.

As you know at 3AM every day a cron script is being run that will search all files on the hard drive and chown some of them (security thing). If I have a file server with lots of small files I do not want such find process to run for those files. Do you know what script runs (by default) and if...

I think you can do dynamic shaping only with ipfw. See this link: http://www.freebsdonline.com/content/view/725/531/ Read Example 5. Router with IPFW Nat Dynamic Limits. That example limits upload to 128Kbps and download to 256Kbps. Instead of hardcoded values, define two variables for upload...

Note that HFSC is kind of linear algorithm, which means if you have lots of queues (like more than 128) it won't scale. I don't know about latest versions of PF but in earlier version they hardcoded HFSC queues to 64 (with a C #define line). On the other hand I don't know how you will do...

You could always use ZFS + Geli.

dinoex@ Your solution worked. Thanks! I've installed it on: uname -a FreeBSD desktop 9.0-RELEASE FreeBSD 9.0-RELEASE #0: Wed Jan 4 15:21:54 EET 2012 me@desktop:/usr/obj/usr/src/sys/GENERIC amd64

Yeah, but Windows 2003 server? Really? :D