[SOLVED/ISOLATED]Re: hfaxd filling my logs with `-> Got a: 127.0.0.1[13] (how to stop this??)

On 10/27/2017 01:53 PM, David C. Rankin wrote:
> On 10/27/2017 11:33 AM, Aidan Van Dyk wrote:
>> Looks like one server is capturing "LOG_DEBUG" to system logs, and the other
>> isn't.  The "Got a: <ip/port>" is a LOG_DEBUG message on in the EPTR command. 
>> There are a smattering of LOG_DEBUG syslogs in the code, those could probably
>> be removed.
>>
>> If you had hfaxd's ServerTracing set to something more chatty (such as 0x33)
>> you would probably see who (network wise) had connected and issued the EPRT
>> command.
>>
>> a.
>>
> 
> Aidan,
> 
>   Thanks for the reply. I've been pulling my hair out on this. I wonder if it
> could be the Sharp Copier (older M355N) I have attached to the network. It has
> a separate fax module (and also supports network delivery of faxes) and it
> also a network scan-to-pdf via a ftp link to a samba share on the server
> (that's all on port 22, so it shouldn't matter). I wonder if it has a 20 sec
> ping or something that is checking to make sure the a network link is up and
> is somehow triggering the logging? I'll have to find the manual and see if any
> of its communications are over the same port. Oh that would be frustrating if
> a dang copier was pinging the hylafax port.
> 
>   Right now, the copier is asleep and I'm not getting any of the annoying
> logs. So I guess the plan going forward is to increase ServerTracing for hfaxd
> to try and identify who the culprit is. I think I have an understanding of
> what to do. When I up ServerTracing to 0x33, will it provide information I can
> identify the culprit with?, or will it be some cryptic bitfield I have to find
> a reference for?
> 

Uugh!

  It's Avantfax triggering the messages! (Mimms...Grrr....) [kidding]. So with:

hylafax 6.0.6-7
avantfax 3.3.5

  When the user logs into avantfax, it has a roughly 20 sec. repeating
communication with hfaxd that is causing the repeating log messages. As soon
as you log out, the messages stop As soon as you log in, the start, e.g.

Oct 31 17:18:44 phoinix HylaFAX[661]: `-> Got a: 127.0.0.1[13]
Oct 31 17:18:44 phoinix HylaFAX[661]: `-> Got a: 54281[19]
Oct 31 17:18:44 phoinix HylaFAX[661]: `-> Got a: 127.0.0.1[13]
Oct 31 17:18:44 phoinix HylaFAX[661]: `-> Got a: 47237[19]
Oct 31 17:19:04 phoinix HylaFAX[676]: `-> Got a: 127.0.0.1[13]
Oct 31 17:19:04 phoinix HylaFAX[676]: `-> Got a: 51127[19]
Oct 31 17:19:04 phoinix HylaFAX[676]: `-> Got a: 127.0.0.1[13]
Oct 31 17:19:04 phoinix HylaFAX[676]: `-> Got a: 55355[19]
Oct 31 17:19:24 phoinix HylaFAX[682]: `-> Got a: 127.0.0.1[13]
Oct 31 17:19:24 phoinix HylaFAX[682]: `-> Got a: 33415[19]
Oct 31 17:19:24 phoinix HylaFAX[682]: `-> Got a: 127.0.0.1[13]
Oct 31 17:19:24 phoinix HylaFAX[682]: `-> Got a: 60561[19]

  It may be worth reworking the default Server Tracing so these messages
triggered by avantfax when a user is logged in are not recorded, and hfaxd not
reset/respawned ever 20 secs. What was filling my logs is I routinely just
stay logged into avantfax with it happily doing its thing in a pinned tab in
firefox.

  Maybe it is an avantfax problem instead? I suspect this is related to how
avantfax is refreshing its display of the current hylafax state. I don't know
how it is refreshed, but is seems like it shouldn't need to trigger a new
hfaxd process every time?

  What say the experts?

cc: mimms

-- 
David C. Rankin, J.D.,P.E.


____________________ HylaFAX(tm) Users Mailing List _______________________
  To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
 On UNIX: mail -s unsubscribe [email protected] < /dev/null
  *To learn about commercial HylaFAX(tm) support, mail [email protected].*