ISTech Support Forum

ISTech Support Forum
http://www.istechforum.com/YaBB.pl
Evo-ERP and DBA Classic >> Problem Reports - Repeatable >> Evo has gone south...
http://www.istechforum.com/YaBB.pl?num=1226328092

Message started by kkmfg on 11/10/08 at 07:41:32

Title: Evo has gone south...
Post by kkmfg on 11/10/08 at 07:41:32

Apparently after I left Friday Evo decided to take an extended vacation as well. We have a 5 user workgroup installation of Pervasive. Evo works perfectly well on the machine hosting the database files. On any other machine Evo will not work properly. I can start it up but most menu applications just hang forever. Strangely it seems that the brand new apps which display the little waiting circling arrow graphic (tas 7 apps) as they load DO work. Ever other program (the older tas 5 and 6 apps) do NOT. People are about to pick up pitch forks and storm the ISTech headquarters... I don't know what else to try... I've tried rebooting the hosting computer and the other computers and it does no good.

Title: Re: Evo has gone south...
Post by GasGiant on 11/10/08 at 08:33:51

Were any Windows updates applied since it last worked? How about patches to security or virus software?

Title: Re: Evo has gone south...
Post by kkmfg on 11/10/08 at 08:36:37

I fixed it... I had changed some things to try to fix some random crashes we get. I just set things back and it worked.

Here is what I had changed:

I turned SPXII off (TCP/IP should always be preferred over this anyway) as a protocol on the workstation server.
I enabled auto reconnect on the workstation server and all workstations (this was because I think some of our problems stem from stale links to the database. Sometimes we get FNUM invalid or 0 errors)

This apparently caused the horrid gutter trash known as PervasiveSQL to silently lock up instead of working or giving some sort of error. But only if I ran tas 5 or 6 apps...

Is there some good reason that tas 5 apps need SPXII or can't have auto reconnect enabled?

Title: Re: Evo has gone south...
Post by kkmfg on 11/10/08 at 10:49:33

And replying to myself once again....

It appears that the setting at fault is pervasive auto reconnect. I actually had SPXII disabled on some of the machines still and it all works so the only setting really changed is auto reconnect. I have no idea why that would lock up the database access.

Title: Re: Evo has gone south...
Post by Lynn_Pantic on 11/10/08 at 21:16:27

Are you using Client/Server or Workgroup? Not sure if it matters but I have had other users enable the TCP auto reconnect without issue so it would be helpful to identify the specifics.

Title: Re: Evo has gone south...
Post by kkmfg on 11/11/08 at 05:51:35

Lynn_Pantic wrote:

Are you using Client/Server or Workgroup? Not sure if it matters but I have had other users enable the TCP auto reconnect without issue so it would be helpful to identify the specifics.

<Gets out a wipe and wipes Lynn's glasses> There.... It's mentioned in my original post and in the sig line of all my posts. ;-) (I'm just giving you a hard time!)

I'm using Pervasive workgroup 5 user. I'd like to know the specifics too! I cannot believe that setting auto reconnect on could possibly do this. However, that seems to be the only change between when it used to work, then when it didn't work, then when it did again. As such I have to believe that it's the cause. I just cannot figure out how. I suppose I could fire up a network packet tracer and do more testing but I'd have to do that on a weekend or after hours. Personally I'd rather just send a 900 pound gorilla to tickle the stuffing out of the people responsible for PervasiveSQL and forget about ever trying auto reconnect.