[nos-bbs] Potential Timer Issue in JNOS (3rd try- sorry)

fish810 fish810 at yahoo.com
Thu Jul 29 11:24:43 EDT 2010


Thanks for the reply, Jack.

Is there any info about how many nodes are having this problem?

Any ideas for a workaround- something better than periodic, automatic restart of 
JNOS?

Regards,
FiSH   K6FSH




________________________________
From: "nos-bbs-request at tapr.org" <nos-bbs-request at tapr.org>
To: nos-bbs at tapr.org
Sent: Thu, July 29, 2010 5:00:03 AM
Subject: nos-bbs Digest, Vol 72, Issue 5


Message: 5
Date: Wed, 28 Jul 2010 13:05:24 -0700
From: Jack Eifer <jeifer at cwo.com>
Subject: Re: [nos-bbs] Potential Timer Issue in JNOS (3rd try- sorry)
To: fish810 <fish810 at yahoo.com>, TAPR xNOS Mailing List
    <nos-bbs at tapr.org>
Message-ID: <4C508D84.5090002 at cwo.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed


Bob,

I've experienced the same problem and I'm using Fedora-Core3. It's a glitch in 
JNOS (and TNOS) which affects its core timer. Most timer related functions start 


going 'negative' and/or report unrealistically large timer numbers. As a result, 


the JNOS console status bar clock appears to freeze and netrom/axip broadcasts 
fail.

It's been my experience that this problem starts after the 25th day of Jnos 
uptime and continues until you reboot Jnos.

Maiko is aware of the problem and looking into it.

Jack, AA6HF




fish810 wrote:
> 
> Hello,
> 
> I have 2 JNOS nodes which each use EXPIRE [interval] to  prune older messages 
> from public areas and AX25 BCINTERVAL [interval] to  set regular beacons.
> 
> Sometime after a long uptime (several  weeks), both nodes have stopped expiring 
>
>
> old messages and both beacons  have stopped.  Same behavior observed on both 
> nodes.
> 
>>From the JNOS console, I get this:
> 
> EXPIRE
> timer: -183/24
> 
> 
> AX25 BCINTERVAL
> Broadcast timer  403528419/1800
> 
> Both functions resumed normally when I manually re-entered proper intervals.
> 
> So  it looks to me like a potential bug in the way some timer register is  
> managed.  Perhaps when the timer expires and gets restarted.
> 
> Has anyone seen a similar problem?
> I'm looking at the timer code but open to  pointers from anyone already 
>familiar 
>
> with it.
> 
> 
> Extra background:
> 
> Both nodes running JNOS2.0h with Ubuntu 9.04
> The original settings were:
> EXPIRE 24
> AX25  BCINTERVAL 1800
> 
> 
> Thanks for any help or ideas you may have.
> 
> Cheers,
> FiSH, K6FSH
> Bob Fishman
>


      




More information about the nos-bbs mailing list