Blogosphere news from you friendly OSS dude.

I was having trouble with an smb mount on my gateway box.
I couldn't cp(1) a file from the smb share to local disk, whereas if I used dd(1) I was able to get the file without problems.
The funny thing was that my laptop runs the same config (freebsd 5.3-stable but from a build of 2 days earlier) and it has no problems with the smb mount.
I couldn't understand why my gateway box was having these problems until I took a closer look at how my LAN NIC was configured.
I had been experimenting with GbE and set the mtu on the LAN NIC to 9000 for some jumbo frame test work and somehow the smbfs implementation (either statically compiled into my kernel of simply by loading smbfs.ko) has problems with this mtu size.
I switched back to the original mtu setting of 1500 and now smbfs works as expected.

posted by Marco @ 22:04 0 comments

Well, this weekend, after having upgraded my laptop from 5.3-RELEASE to 5.3-STABLE using SCHED_ULE + PREEMPTION, I have now also upgraded my home gateway/router/packet-filter box to 5.3-STABLE as well.
Earlier, I used webbench from my laptop (webbench -c 100 -t -1 (100 processes)) to stress-test the gateway box running Apache2.
Running webbench with these options, system load on the gateway box would rise to ~120 and the system felt a little sluggish when using the vi editor for example.
After having made the switch to SCHED_ULE + PREEMPTION the box feels snappier even at the same system load.
However, I use the munin and munin-node ports on my gateway box and I do admit that the munin-node process (perl) refused to update any of the statistix while the system was running under a >100 load.
Only after having killed the webbench stress-test the statistix were gradually updated (or at least, the generated png's were).
This is something I need to take a closer look at.

posted by Marco @ 22:06 0 comments

I run FreeBSD 5.3-STABLE on my gateway/router/packet-filter box.
Currently it's using the SCHED_4BSD scheduler.
I read on the -STABLE mailinglist that the ULE scheduler was available for testing once more for RELENG_5.
As a result, I did the full 'make world and kernel' on a 5.3-STABLE workstation and it's running happily with SCHED_ULE since 2 days now.
If I'm happy with its performance and stability I'll make the switch to SCHED_ULE on my home gateway box too.

posted by Marco @ 21:02 0 comments

Well, I saw Blade: Trinity last night.
For a single movie, it was alright but compared to the previous 2 installments, it left me with an unsatisfied feeling.
I found the first blade film to be, ofcourse, original and also a good and dark film.
IMO, the second film had, like many sequels, lots more action but had a more shallow storyline.
Unfortunately the third film was missing both elements which made the previous 2 films so good.
To say blade 3 was a dull film would be inaccurate but the chemistry between the characters was somehow missing.
I am dissappointed to see the trilogy being ended this way but there's nothing I can do about this.
When the movie will come out on DVD I'll probably buy it to complete the trilogy but I had simply hoped for a better ending of some cool stories of a cool character.

posted by Marco @ 21:10 0 comments

In just a couple of minutes I'm off to the movies to go see Blade: Trinity.
I'm hoping the movie is as good as the previous 2 installments.

posted by Marco @ 21:51 0 comments