• Welcome to OS2World OLD-STATIC-BACKUP Forum.
 

News:

This is an old OS2World backup forum for reference only. IT IS READ ONLY!!!

If you need help with OS/2 - eComStation visit http://www.os2world.com/forum

Main Menu

Time out issue : is there a way to remove drives no longer active ?

Started by microdrome, 2008.04.02, 22:12:26

Previous topic - Next topic

microdrome

Using peer ,netdrive for years with no special problem - even when connecting to various winXX machines-
still leaves me with one issue :
When a computer in my network is switched off or the lan cable removed is does not automatically ejects the drive.

Is there anyway to let this happen ?
Some programs which scan the drives at startup or opening the file dialog creates a huge time problem which holds the system.
This can take up to minutes before the program filnally sees that drive "x" is no longer available.
Perhaps a time setting somewhere ? or a deamon which refreshes the drives at an interval (such as with USB insert/removal)

Curious if others have the same proble and did not get around solving it.

*ps even winXX programs do not have this annoying behavior :'(

warpcafe

Quote from: microdrome on 2008.04.02, 22:12:26
When a computer in my network is switched off or the lan cable removed is does not automatically ejects the drive. Is there anyway to let this happen ?

Hi microdrome,

not sure as I don't have an eCS/OS2 machine around at the moment... so this is purely guessing:
Check the ibmlan.ini entries "wrkheuristics" (IIRC). There is a whole bunch of flags available along with a short help text what they mean. Perhaps there is some parameter available which does the required trick... tell me if that helped to some extent, otherwise I'll give it another look when I'm back home.

regards,
Thomas
"It is not worth an intelligent man's time to be in the majority.
By definition, there are already enough people to do that"
- G.H. Hardy

microdrome

 :( checked the ibmlan.ini file but i cannot see anything which makes sense there - there are some timeout values for netbios ,but that cannot be it since it is set to 35 sec.
Besides the same effect happens also on a "netdrive" drive and this one has nothing to do with the os/2 peer

It must be inside os/2/ecom  itself..


too bad....
Peter

warpcafe

Quote from: microdrome on 2008.04.03, 18:12:30
there are some timeout values for netbios ,but that cannot be it since it is set to 35 sec.

Ok, regardless of the fact that that Netdrive shows the same problem... remember that the timeout is per access... so if there is a retry happening or a different polling right after the 1st timeout, you will have 35 secs. AGAIN - leading to roundabout 1 minute (70 seconds) of hourglass-watching.

I will check anyway what the cause may be... BTW, what OS and version are you running - still the same one like throughout all the years without problems (as mentioned) or have you recently upgraded? I mean - either OS-upgraded the machine in question, or perhaps a new Win-machine on the network?

Cheers
Thomas
"It is not worth an intelligent man's time to be in the majority.
By definition, there are already enough people to do that"
- G.H. Hardy

microdrome

I am running ecom 1.2 ,2.0 rc4 ,some beta of warp4 and somtimes 1 or 2 win2000 or xp  machines.
There has always been this problem in fact ,but it was not a big deal in the past.
Ever since i turn machines on and off (power saving $$) it began annoing.
+ a portable NAS which is not always on.  (and netdrive keeps the mount point active wheter the drive is available or not )
I used to leave computers on 24/7 ,but they are consuming too much power nowadays and the bill gets bigger.

So i have this behaviour sice i began using os/2 1.2 ,only now i finally am trying to seek a solution.

Thanks for the reply's so far.
peter

RobertM

The problem may have nothing to do with NetBIOS (see last paragraph) - but instead is possibly something to do with the WPS replacements in eCS.

IIRC, under Warp and WSeB (no WPS enhancements), once a connection share is enabled, either two things happen when the connection is broken (and the client becomes aware of such)... usually it is the second.


  • (1) The drive icon dissappears (usually requires a refresh or an attempt to access the drive)... rare I have seen this happen otherwise
  • (2) The drive icon changes and the ready light on the icon turns red. (under the Lan Server Admin Tools, certain objects do not dissappear either, but becomes "greyed out" - for instance server representations).

I have a feeling that there may not be a suitable secondary icon state for an inactive/unavailable share in the WPS enhancements - either that, or the various "folder-refresh" enhancements may not be reading that info...

This is just a guess from memory from a long while back - and either (a) may not apply, and/or (b) may not be accurate (as I have been running one form or other of WPS enhancements on most of my machines for a while).

-Robert


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


Andi

Maybe I'm wrong, but I think it's not wps realated. At least not directly. I think these are netbios timeouts and maybe tcpip timeouts. You can change in ibmlan.ini netbiosretries and netbiostimeout to lower values, but unfortunatly very low values do not work. I don't no the limits. Unfortunatly to low values are silently ignored. Together with the retries and a lot of individual trial accesses, it takes rather long until the system knows that the ressource is unavailable.

To make things really worse, the wps (and probably your programms) seem to hang during this unsuccessful accesses.

Same thing with ftp ressources which have nothing to do with netbios but tcpip. I usually access ftp-ressources via netdrive and LarsenCommander. When these ftp ressources are not available then LarsenCommander 'hangs' but not the whole wps. With ftp-ressources over netdrive you can set retries and timeouts too. But these are 'on top' of the standard tcpip timeouts. These are probably tuneable too, but beyond my knowledge and probably have a negative effect on other things.

So the current situation is disappointing for me too. Maybe samba  as a replacement for netbios can help, but I didn't try till now.