Lately I had a lot of Vera startups, sometimes every 30 minutes. I had a look at the logs to see if I can find anything. This is from just before the lua crash:
[7;36m2014-10-06 20:47:35 - LuaUPnP Terminated with Exit Code: 138 [1;00m
[7;36m2014-10-06 20:47:35 - LuaUPnP crash [1;00m
01 2014-10-6 20:32:20 caught signal 11 <0x2d797680>
---------------exited-------------
I also found a few other things in my logs which do not look right, and might indicate a problem:
09 10/06/14 20:47:37.237 UserData::ParseRooms 13 Cameras <0x2b026000>
09 10/06/14 20:47:37.237 UserData::ParseRooms 14 AV <0x2b026000>
03 10/06/14 20:47:37.326 UserData::LoadUserData BuildVersion: 1.5.622 SvnVersion: 7546 Model: 31 Sercomm NA900 flush: 0 changed: 0 resync: 10065 syncdevice:0 <0x2b026000>
02 10/06/14 20:47:37.349 [33;1mJobHandler_LuaUPnP::FetchMissingFiles need to download plugin 188 [0m <0x2b026000>
02 10/06/14 20:47:37.350 [33;1mJobHandler_LuaUPnP::m_sMissingFile_set plugin [0m <0x2b026000>
02 10/06/14 20:47:38.717 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.721 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.725 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.729 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.730 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:MiosUpdater1 [0m <0x2b026000>
02 10/06/14 20:47:38.734 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.734 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.737 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId:ZWaveDevice1 [0m <0x2b026000>
02 10/06/14 20:47:38.743 [33;1mDevice_Service::MatchupUpnpServices no upnp service for upnp-rfxcom-com:serviceId:rfxtrx1 [0m <0x2b026000>
02 10/06/14 20:47:38.743 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:rfxcom-com:serviceId:rfxtrx1 [0m <0x2b026000>
02 10/06/14 20:47:38.743 [33;1mDevice_Service::MatchupUpnpServices no upnp service for urn:schemas-rfxcom-com:device:rfxtrx:1 [0m <0x2b026000>
02 10/06/14 20:47:38.744 [33;1mDevice_Service::MatchupUpnpServices no upnp service for upnp-rfxcom-com:serviceId:rfxtrx1 [0m <0x2b026000>
and this goes on for quite a while. The there is something else:
My Vera is crashing also (middle light goes out, which I assume is LUA/LUUP no longer responding). I put an automated luup.restart() scene that ran twice a day. Didn’t work. The crashes got more frequent until now it appears to stay up less than an hour…
Boy, I would love to solve this one. I sent mail to the support folks; is a phone call required?
I can find the Signal exits in the log file, but no clue what change might have caused this. I removed all the add-ons that might have caused an instability.
I am running UI7 and the latest firmware; I’m asking for trouble I know…
Quite suddenly, a few days ago, after a power cycle my VeraLite refused to work. Its web page tells me that the Lua engine took too long to start. Of course, in clasic MicasaVerda fashion, it doesn’t give me any explanation nor does it provide actions which I can take. THey have been silent for several days to my request for tech support.
I SSH’ed in and grabbed the logs. They give me something similar:
…then there are several entries with Device_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId for a dozen or so devices.
For all practical purposes this unit is now a brick.
My questions:
a) Is this indicative of Vera’s quality? Should I just give up on this and go with HomeSeer or something more reliable?
b) WTF? How do I remedy this?
[quote=“mrslother, post:3, topic:183437”]Quite suddenly, a few days ago, after a power cycle my VeraLite refused to work. Its web page tells me that the Lua engine took too long to start. Of course, in clasic MicasaVerda fashion, it doesn’t give me any explanation nor does it provide actions which I can take. THey have been silent for several days to my request for tech support.
I SSH’ed in and grabbed the logs. They give me something similar:
…then there are several entries with Device_Service::MatchupUpnpServices no upnp service for urn:micasaverde-com:serviceId for a dozen or so devices.
For all practical purposes this unit is now a brick.
My questions:
a) Is this indicative of Vera’s quality? Should I just give up on this and go with HomeSeer or something more reliable?
b) WTF? How do I remedy this?[/quote]
I fixed this by swapping a config file that was backed up 3 minutes prior to the configuration becoming unstable/corrupted. BTW I contacted support to help but have not heard from them in few days so I fixed it by following directions that RichardTSchaefer posted in another thread.
cd /etc/cmh
cp user_data.json.lzo.3 to user_data.json.lzo
As for the crashes I see them too and it is mentioned in another discussion that the memory allocated to the LuaUPnP process is over 100%, mine is around 130% - you can check this by running the TOP command.
I am testing the Vera setup alongside my Incontrol PC software but so far I am not impressed with the stability. The only reason I even tried Vera is the PLEG plugin which allows for extended logic.
[quote=“hmspain, post:2, topic:183437”]My Vera is crashing also (middle light goes out, which I assume is LUA/LUUP no longer responding). I put an automated luup.restart() scene that ran twice a day. Didn’t work. The crashes got more frequent until now it appears to stay up less than an hour…
Boy, I would love to solve this one. I sent mail to the support folks; is a phone call required?
I can find the Signal exits in the log file, but no clue what change might have caused this. I removed all the add-ons that might have caused an instability.
I am running UI7 and the latest firmware; I’m asking for trouble I know…[/quote]
Quick update. I found I had WAY more devices than the VeraLite was intended to handle. I upgraded to a Vera3 and now my system runs stable. You hate the thought of spending for yet another controller, but think of your investment in Zwave… the controller is only a small (albiet important) part.
Yes i have the nightly heals turned on. But it has been for months now. Unless there are other heals running very occasionally, which is not something i am aware off. I am not so sure if it is a good idea to have the nightly heals running. I think once my system was running nicely i turned them off. Not sure why i turned them back on again.
It is good to get to a point of stability (I know…) where the changes come less frequently, and you can isolate a single change as the cause for a problem…
Let Vera decide on Zwave routing, let Zwave decide on Zwave routing? To heal or not to heal? Many times I go back and forth; hopefully learning something in the process.
I hear a lot of complaints from folks just getting started (or not giving up quickly) that Vera is too difficult to manage. Well, you learn, it’s not plug and play! Rather than laying fault, you realize that the system is quite complex. It may be a Zwave device problem, or a network problem, or a UI problem with Vera, on and on.
My Vera 3’s LUUP process is crashing about 20 times a day! Vera does restart the LUUP process, but I often wonder if my slow motion detection response is just the LUUP process restarting, or something else :-).
I’m not sure if I have a corrupt configuration, but the user_data file does get updated every crash. It keeps getting bigger and bigger, but not large enough to worry me about disk space. Sure wish I could do a diff on these files but the user_data file is binary so I can’t really see what is going on.
I don’t think I have a corrupt configuration (the Vera runs fine between restarts). I’m tempted to try doing a backup, copying the original (small) user_data file into production, reboot, and restore… but I’m curious if these steps might work in my case.
[quote=“RichardTSchaefer, post:11, topic:183437”]Actually the file is not binary … it’s a compressed JSON (ascii) formatted file.
You can use:
pluto-lzo d SrcFile TargetFile
to decompress.
Then diff … Depending on the configuration the files can be large and the changes spread out throughout the file.
Your strategy will not hurt … and may solve a problem is something was corrupted.
Have you changed anything lately ? Add a new plugin or device ?[/quote]
[quote=“hmspain, post:10, topic:183437”]My Vera 3’s LUUP process is crashing about 20 times a day! Vera does restart the LUUP process, but I often wonder if my slow motion detection response is just the LUUP process restarting, or something else :-).
I’m not sure if I have a corrupt configuration, but the user_data file does get updated every crash. It keeps getting bigger and bigger, but not large enough to worry me about disk space. Sure wish I could do a diff on these files but the user_data file is binary so I can’t really see what is going on.
I don’t think I have a corrupt configuration (the Vera runs fine between restarts). I’m tempted to try doing a backup, copying the original (small) user_data file into production, reboot, and restore… but I’m curious if these steps might work in my case.[/quote]
No guts, no glory…
I made a current backup of my Vera 3 (and uploaded it to the server just in case)
I renamed all the /etc/cmh/user_data.json.lzo.n files to save them
I copied the original /mios/etc/cmh/user_data.json.lzo file to /etc/cmh
Rebooted (Yikes, I have a new Vera 3!)
Did my first restore from backup crossing my fingers
Rebooted
The restore worked (I know, you guys do it all the time…)
Looking over the new /etc/cmh/user_data.json.lzo files, I can’t see much difference. If my Vera stops restarting the LUUP process multiple times a day, I’ll declare victory. Stand by…
No good; it is still restarting several times an hour. Support promised to call me Monday afternoon. I’m hopeful they will see something. I have the log file set on verbose.
Best Home Automation shopping experience. Shop at Ezlo!
