-
Weird Network Problem with HWQS
Hello there,
I'm having a very very weird network issue with a multi processor HWQS project. When i go to the activation screen, all processors are found and communicate properly.
but, when i go to the transfer screen, it won't communicate with any processor there.
I have multicast enabled on my entire network, across multiple switches. However, if i choose to connect using unicast, by select one of the processors ip address, this happens:
While, at the same time, if i choose other ip from that same list:
Do you guys have any idea of what may be wrong here?
-
Forget multicast.
It looks like all of your processors are in separate panels. With unicast 2-4 upload ok so I would check the wiring on processor 1. Multicast is flakey but maybe it is able to fix something with processor 1 that unicast cant.
-
Originally Posted by
randyc
Forget multicast.
It looks like all of your processors are in separate panels. With unicast 2-4 upload ok so I would check the wiring on processor 1. Multicast is flakey but maybe it is able to fix something with processor 1 that unicast cant.
Yes, each processor is in a separate panel, across multiple fiber uplinks and switches. The same wiring was worked before and was checked and re-checked numerous times, so it's not a wiring issue. The weird thing here is that using unicast i can connect directly to each processor, while multicast, that is like a simulated multiple unicast, where the transmitter sends data to the multicast group and each listener that joins the multicast group receive data at the same time. I have IGMP enabled on each and every switch that traffic goes by, multicast routing enable, but for some reason traffic doesn't reach one of the processors.
- - - Updated - - -
Originally Posted by
randyc
Forget multicast.
It looks like all of your processors are in separate panels. With unicast 2-4 upload ok so I would check the wiring on processor 1. Multicast is flakey but maybe it is able to fix something with processor 1 that unicast cant.
Yes, each processor is in a separate panel, across multiple fiber uplinks and switches. The same wiring was worked before and was checked and re-checked numerous times, so it's not a wiring issue. The weird thing here is that using unicast i can connect directly to each processor, while multicast, that is like a simulated multiple unicast, where the transmitter sends data to the multicast group and each listener that joins the multicast group receive data at the same time. I have IGMP enabled on each and every switch that traffic goes by, multicast routing enable, but for some reason traffic doesn't reach one of the processors.
-
Could you repeat that? (just kidding!)
I'm sure you checked the wiring multiple times. Try swapping processor 1 and one of the others. I'm not 100% sure what that would prove but if the issue follows the processor it would seem to indicate an issue with the processor. If it stays with panel, it would seem to indicate an issue with the network.
-
Not sure if this helps, but the other day I had a brand new HQP6-2 that I had to activate and load a 13.0 program to. I was first trying remotely via VPN and I was able to activate it by setting the system communication to the processors reserved IP instead of Multicast. But After that I could not transfer successfully. I also could not connect with Terminal, or update firmware. When on the activation screen the processor would show activated. I checked the processor from the ip address and saw it was on SW 9.1.6. So I call support and of course they do not support VPN connections (what year is this really???) and aside from the usual garbage suggestions of turning off my firewall and antivirus they were of no help. Also told them the unit was on 9.1.6.
So I tried teamviewer to an on-site technician and after installing 13.0 on his machine I ran into the same exact problem. This is ridiculous as I am now on site plugged into the same network. I then saw in some QS release notes that 13.0 will not allow terminal connections to older FW processors. So I opened 11.5 and connected to the unit and was able to update the firmware to 11.5. From that point I was able to update to 13.0 remotely and load my program in. Why are they shipping processors with ancient firmware that you can not connect to with new software??? I've seriously had it with Lutron between the transfer issues I've always had and the newer telnet lockouts that kill all my Savant/Crestron integrations. How did support not catch the firmware mismatch especially when I asked the rep about the old FW???
Sorry for the mini-rant but maybe the firmware/SW mismatch issue I had may be similar to what you are seeing.
-
Hello sergiobaiao,
Thank you for your post.
Send a support file to hwsupport@lutron.com and we can escalate the issue. We will likely need some wireshark captures in order to further investigate. Our tier 2 technical support team can get you instructions on how we need to setup the captures.
Hello johnmax,
Thank you for your input on this topic and your feedback. I will ensure this is reviewed by the appropriate teams.
-
Originally Posted by
Blair S.
Hello sergiobaiao,
Thank you for your post.
Send a support file to
hwsupport@lutron.com and we can escalate the issue. We will likely need some wireshark captures in order to further investigate. Our tier 2 technical support team can get you instructions on how we need to setup the captures.
Hello johnmax,
Thank you for your input on this topic and your feedback. I will ensure this is reviewed by the appropriate teams.
I can't even create the support file succesfully. How should i contact the tier 2 team?
-
Post Thanks / Like - 0 Thanks, 1 Likes, 0 Dislikes
-
while playing with wireshark today, i've found something both interesting and strange at the same time...despite the lutron software defaults the multicast group to 239.0.38.1, when i started a transfer on Lutron Designer, it sent multicast packages to 224.0.37.42 and the processors replied on this multicast group. I'll be adding this group to my switches config and see if that works.
-
Post Thanks / Like - 0 Thanks, 1 Likes, 0 Dislikes
-
Had the same problem today as in my above post. Brand new HQP6-2 with 9.1.6 firmware on it. Again it activated in 13.0 but could not terminal into it or transfer. Now I open up a 11.5 file and I can terminal into it no problem, but in the activation screen it finds it 1 out of 5 times. This is with using either multicast or directly to the processors IP. I had to try about 20 times to update the firmware to 11.5 so I can open 13.0 and update to that and get my program on there. And I am wired into the network the processor is on. The same switch next to my desk!!! Why why why is it so difficult to get a processor up and running? These connection issues are ridiculous, and why are we still shipping processors with 2 year old firmware that we cannot connect to at all with current software?
-
Post Thanks / Like - 1 Thanks, 0 Likes, 0 Dislikes
-
[QUOTE=johnmax;25521]Had the same problem today as in my above post. Brand new HQP6-2 with 9.1.6 firmware on it. Again it activated in 13.0 but could not terminal into it or transfer. Now I open up a 11.5 file and I can terminal into it no problem, but in the activation screen it finds it 1 out of 5 times. This is with using either multicast or directly to the processors IP. I had to try about 20 times to update the firmware to 11.5 so I can open 13.0 and update to that and get my program on there. And I am wired into the network the processor is on. The same switch next to my desk!!! Why why why is it so difficult to get a processor up and running? These connection issues are ridiculous, and why are we still shipping processors with 2 year old firmware that we cannot connect to at all with current software?[/QUOTE
Please consider that some of us don't run the most current firmware. I would be very disappointed if Lutron started shipping processors with 13.0 loaded- you can't roll those processors back and I'm generally using 11.5 for most of my current projects. I don't update the firmware unless I need a hardware or software feature.
I totally sympathize with your rant. I have a couple of large projects with 4+ processors and experience weirdness with LAN connections to the processors when onsite, when on VPN, when wireless, when wired. I am absolutely clueless as to why a company as successful as Lutron can't have as good or better LAN coms than the other big boys like Crestron and Savant.
-
Post Thanks / Like - 0 Thanks, 1 Likes, 0 Dislikes