Rebooting controller frequency

[dentalben]

Just wondering how often people find they have to hard reboot (pull power, plug back in) their controller(s)?

I have an HC300B as my main controller with an HC200C as the secondary and I find that about once a month I have to do a hard reboot because things in my house just don't work very well (keypads are very delayed (from several seconds to almost a minute), screen saver on IE touch screens don't work, etc). Doing the reboot gets everything working again, but it's a hassle and takes 30+ min to get everything back online again.

[ILoveC4]

It shouldn't take anywhere near 30 minutes to get back online.

I think what you're referring to is a well known, and well documented zigbee bug. I have my stuff setup to automatically reboot every two weeks in the middle of the night, until this bug gets fixed.

[dentalben]

While some of my items are "workable" in about 15 minutes, it does take about 30 to get all of my switches, etc to communicate again. I'd love to know how you configured yours to reboot automatically as I'd love to set mine up to do the same.

[dbuono]

How do you set your main controller to reboot in the middle of the night? I run into this zigbee slow down problem every two to three weeks. The reboot in the middle of the night is an excellent idea.

[ILoveC4]

How do you set your main controller to reboot in the middle of the night? I run into this zigbee slow down problem every two to three weeks. The reboot in the middle of the night is an excellent idea.

I have the power strip programmed to cut the power to certain ports.

[CFUG]

Same here- web switch. It's getting bad here- about twice a month is needed. 30 minutes? I'm fully functional in a few minutes.

[Tokead]

Same here- web switch. It's getting bad here- about twice a month is needed. 30 minutes? I'm fully functional in a few minutes.

I'm running 2.06 on a HC1000 & 300 with no switches. In a recent power outage it took approx' 5 mins to get basic control but about 20 - 30 mins for full functionality. My dealer advised that it'll take about 20 mins for the 300 to get up to speed. (the project is on the 1000 which is on a UPS)

[EagleMoon]

Does this bug only affect certain kinds or versions of hardware? I don't think I've ever experienced it but maybe it's because my system is small and I only live there about 60% of the time.

However, my HC300 has now been running 172 days and I haven't seen any problems.

[ILoveC4]

Does this bug only affect certain kinds or versions of hardware? I don't think I've ever experienced it but maybe it's because my system is small and I only live there about 60% of the time.

However, my HC300 has now been running 172 days and I haven't seen any problems.

It's a 2.0.x bug.

[EagleMoon]

It's a 2.0.x bug.

Whew! Thanks.

Why do people keep saying that 2.0.6 is solid? Having the system become unstable and needing to be rebooted every two weeks does not fit my definition of a solid, mostly bug-free release.

[CFUG]

^Who is saying that? You should know that no build addresses all possible fault paths.

[CFUG]

Same here- web switch. It's getting bad here- about twice a month is needed. 30 minutes? I'm fully functional in a few minutes.

I'm running 2.06 on a HC1000 & 300 with no switches. In a recent power outage it took approx' 5 mins to get basic control but about 20 - 30 mins for full functionality. My dealer advised that it'll take about 20 mins for the 300 to get up to speed. (the project is on the 1000 which is on a UPS)

Don't know what to tell your dealer- mine is good-to-go inside of 5 mins.

[ILoveC4]

^ mine too!

[EagleMoon]

^Who is saying that? You should know that no build addresses all possible fault paths.

Uhhhh.... as just one example, maybe the guy on post #13 here:

http://c4forums.com/viewtopic.php?id=7993

And yes, I know about shipping product with known bugs -- I used to work in, and later managed, a system test group for product development. We tried to classify bugs by severity, with the most severe group requiring resolution before shipping. What I was saying is that a bug that is commonly encountered in the field and requires a reboot every two weeks would have been in that class for me. However, I can easily imagine some scenarios where releasing it was a reasonable decision for Control4. But note, that's NOT the same as agreeing that it's a solid release that everyone should install.

[ILoveC4]

Control4 didn't know about this bug for a while, because it takes several weeks to manifest itself. Development was so rapid and versions wee being updated so quickly (especially beta) that most systems were getting updated (which involves a reboot) before the bug appeared.

[EagleMoon]

I can understand that.

One of the scenarios that I imagined is that the bug was thus discovered sometime just before the 2.0.6 release but was going to take awhile to correct, yet there were a number of other very serious bugs whose fixes were ready to go around the time it was discovered. So those fixes were released as 2.0.6, knowing this other one still existed. If that were the case and I'd been in the meetings, I would have agreed with releasing what we had ready to go. Pointing out known and unresolved problems is part of what release notes are for.

And anyone who was *already* running a version of 2.0 should probably be encouraged to upgrade.

My disagreement was with whether it should be characterized as "pretty solid", "with most the major problems resolved" and to which everyone should consider installing, even those of us on pre-2.0 systems. But those comments are generally what I heard several times here and from my dealer.