Lockup's
i have a problem.
i have 3 jogglers all running xapflash.
sometimes (once every 2-3 days) they stop responding to commands a restart of joggler does not resolve problem.
if i go to we pagei can then control rf devices so HAH is sending out commands correctly, even xapflash on webpage does not respond but from admin it works fine.
a reboot solves problem or if i wait for 10-20 minutes problem seems to be resolved.
i have run xfx viewer and all looks fine BUTno rf on/off is detected whilst problem is exhibited, also if i swith rf on/off several times whilst problem is occuringsaiditem will within 5-10 minutes switch on/off a corresponding number of times.
1 how do i debug this
2 anyone else with joggler/xapflash noticed this?
driving SWMBO nuts
oh and a rebootof HAH instantly fixes it.
Dean
Oh so the joggler MAY be queuing them up and not sending them - that would be an xapflash issue then. Pester Kevin :)
Or a transmission onto the wire from the iServer - pester me. Given each joggler misbehaves the same way is sounds more like my problem. :(
Do try putting a local iServer on one of them and see how that goes --- NOTE: the version on the joggler is from an older code base. This code is what you currently have with 306 but I've not updated the joggler binary with the changes I made for 307.
http://www.dbzoo.com/livebox/xapflash#iserver_installation
This should at least isolate any iServer issue to a single joggler - I would update with 307 before doing this.
Brett
Dean... did you fix this ?
I don't believe this is a xAP Flash issue but you never know... it's not one that has cropped up elsewhere or that I experience in a similar setup. I assume the Joggler still shows as connected to the iServer ? Is this WiFi or wired ? Do remote changes in any xAP devices update their status on the Joggler instantly still ?
The messages between the Joggler and the HAH iServer are passed via TCP so you don't see those in xFX Viewer until the HAH iServer pushes them out as UDP. As such it is difficult to ascertain just where they are being queued (Joggler or HAH).
Another diagnostic would be to temporarily run the Windows based iServer application and see if the problem remains, if so then it's my issue. If you are familiar with Wireshark that would help diagnose it too.
K
Dean just reading about your problem. So you are saying that your jogglers don't reponds to commands but you know the HAH is send these out as xfx viewer see them on the wire.
Rebooting the HAH resolves the issue!? I would say this is happening.
An xAP message is being sent out of from the HAH via the xap-hub and this is what you are seeing and saying..
BUT the jogglers get their messages via the iServer, and this is receiving its message also from the xap-hub so it should be getting the message. Perhaps its jamming trying to send it out back as this would affect all 3.
I take it you are using a single iServer on the HAH. What you can try is installing a local iServer on each and point their xapconfig.xml into 127.0.0.1 instead. At least this way a failure of iServer (if this is what it is) will not take out all 3.
Alternatively update to the 307 build (if you can) as I made a number of iServer tweaks with regards to messages being lost. Transmission to each joggler now happens in a separate thread from the receiver side thread. This should alleviate issues with messages being lost whilst processing outbound packets.
Brett