[fonosfera] Some early RC2 torrent testing

Deco decodecoding at gmail.com
Thu Jun 4 09:38:40 UTC 2009

I agree this report. Same happens to me; after a certaim time downloading with torrent,  wifi link to client is interrupted and although private network SSID can be seen by clients, association is not possible.
Beside this, a logread -f will show chillispot restarting over and over.

Other minor issue is that the file downloaded is not placed inside the torrent folder but top directory..



-----Original Message-----
From: Jorge Suárez de Lis <yo at jorgesuarezdelis.name>

Date: Thu, 4 Jun 2009 10:47:42 
To: <development at fonosfera.org>
Subject: [fonosfera] Some early RC2 torrent testing

[ Note: This is an adaptation from a blog comment I posted on
http://blog.fonosfera.org/long-time-no-see-rc2-is-here/#comment-4236 ]

Hello! This is the first version I test :)

I'm trying to download a torrent. From time to time I'm kicked from
the private network and I can't reassociate until I wait some minutes.
 When I can finally reconnect, the torrent has been paused. I can
unpause it, and the same goes once and another.

I've attached a dmesg output. http://pastebin.com/m7f9998c6 and I'm
worried about three things:

1. That WARNING at line 180.
2. NETDEV WATCHDOG: wifi0: transmit timed out
3. Constantly remounting the hard drive?

I fsck'ed the partition and I am running some mor testing at home.
Maybe some of these messages are normal, others are not, and could be
caused by the bad unmounted partition. But I actually don't think so.
My guess is:

1. Unworkable eats all the ram memory
2. Because of that, the fonera becomes unresponsive for a while and kicks me.
3. After some time, the system terminates the unworkable proccess, so
the torrent pauses, ram is freed and i can connect again.

This is only a guess. I'll have to run some more tests. How can I know
the amount of ram unworkable is taking? vmsize also takes in count the
mmaped memory so top and ps gives me an inaccurate amount.

If this continues to happen, perhaps I should create a swapfile on the
hard drive and see what happens next. If that solves anything, it
wouldn't be very traumatic to create some temporary swapfile on the
hard disk, would it? But perhaps I'm guessing too much only from these
tests, so I'll keep watching what happens by now :-)

Jorge Suárez de Lis <yo at jorgesuarezdelis.name>

Development mailing list
Development at fonosfera.org

More information about the Development mailing list