Hopper daily reboot annoyance

[neonturbo]

Why does my Hopper insist on rebooting every day at 1:39 AM? I usually get the first "check for update" prompt at about midnight, (I select no at that time) and at 1:39 it simply reboots no matter what without a prompt. Very annoying when you are watching TV to have it reboot in the middle of a program. :mmph:

I have tried every update setting in the [menu>settings>diagnostics>updates] menu. It ignores every option. This is is a new thing that has happened since I got S238 a week or so ago. I have tried multiple power cord reboots with no change.

Any ideas?

 

[KKlare]

Just about every Dish receiver does reboot daily. The Hopper may be the first to not allow a time shift.
I used to move the time on 622/722 and before later and earlier to miss its reboot, but...
The reboot is the "lazy" way to fix memory leaks--not returning memory after allocating it.
If you did not do a reboot within 2 days on the 722, likely your machine would reboot for a bad stack.
The Hopper is likely the first to check for scheduled recordings and wait a little while.
I have not had it reboot during a recording but it is possible or reboot may extend into the start time.
My computer's MacOS manages to run for months without one--minimal leaks. Better programming.
It may ignore your (just) viewing because it does not know if you are really there.

-Ken

 

[KKlare]

To see the really important time, the file system check or fsck, look into Settings>Network Setup>Tools>View Counters.
Then page down many times to the fsck date. If that is more than a few days ago you are looking for trouble.
-Ken

 

[dwarren2]

It is also the way it downloads a updated epg and any new software. If you are a late night person, set up a recording that runs till after you go to bed and delete it the next day. That way the reboot won't happen till after you go to sleep.

 

[Hall]

Dish must have a very unusual method of updating the EPG if it requires a reboot. I've seen OTA channels update on-the-fly after scanning for channels or on older receivers, if you scrolled into the future far enough, it would add data, so clearly a reboot isn't mandatory.

Software updates - how often does that happen? A reboot isn't needed to get the update, only install it.

 

[dwarren2]

Tell you what. Why don't you call up Dish and tell them they don't know what they are doing. Why don't you rewrite the software so it does what you want.

 

[Hall]

Ooooh, someone's a bit sensitive....

 

[3HaloODST]

All of this would be solved if they'd just allow us to change the reboot time.

Sent from my iPhone 4S using Forum Runner

 

[John Kotches]

To see the really important time, the file system check or fsck, look into Settings>Network Setup>Tools>View Counters.
Then page down many times to the fsck date. If that is more than a few days ago you are looking for trouble.
-Ken

That, frankly is a load of hooey.

I have systems with a lot more than a few days on them that I have no concerns about an fsck on my systems and I have quite a number of them on varying platforms. This is especially true once you go to a journaled file system such as ext3fs or ext4fs (there are others, but we'll stay on topic). I think they have the fsck interval set too short. I know I accept the default of 180 days, i.e. if it hasn't had an fsck in 180 days, it will do one at next mount and I think that might be too aggressive. The amount of time it takes to fsck these file systems adds up, especially if they are doing these serially rather than in parallel.

Caveat, if it's been up for more than a year, and yes I have a number that have been, I get nervous about a reboot in general.

Have you spent any time administering any *nix based systems with journaled file systems? Doesn't matter which one, could be HP-UX, AIX, Linux, Solaris, OpenBSD etc.

 

[John Kotches]

All of this would be solved if they'd just allow us to change the reboot time.

Sent from my iPhone 4S using Forum Runner

If there are no changes, there shouldn't be a mandated reboot. Plain and simply put if their software stack is so unstable that it requires a daily reboot then there's much bigger problems that I would be concerned about.

 

[Hall]

I do have to agree that it seems that the most likely, but unofficial reason, is memory leaks/instability. Myself, and others, really noticed that on the 622, forcing a reboot every week (or more often) made for a much smoother running system.

 

[dwarren2]

If there are no changes, there shouldn't be a mandated reboot. Plain and simply put if their software stack is so unstable that it requires a daily reboot then there's much bigger problems that I would be concerned about.

That would make programming the software even more difficult. Right now they just have to flag the range of receivers that will get software updates each night. When the reboot happens it automatically downloads. In your secenerio, they would first have to send a command to each receiver telling it to reboot, then it would reboot. Just another point where things could get fouled up. But I do agree that the option as to set your own time for the reboot was nice when I had my 722. But I don't know if I've lost any recordings when I've set a timer for the middle of the night. Just this weekend, I had timers for all the F1 shows from Malaysia on NBC Sports(159). None of them were missed due to a reboot.

 

[KKlare]

That, frankly is a load of hooey.
I have systems with a lot more than a few days on them that I have no concerns about an fsck on my systems and I have quite a number of them on varying platforms. This is especially true once you go to a journaled file system such as ext3fs or ext4fs (there are others, but we'll stay on topic).

I appreciate your greater experience with these file systems, but....
Well, I find you are wrong about the ability to have fragmented files on Dish's version of ext systems.
I cannot fit files if there is not enough contiguous space, even with lots of total free space.
It seems, to me, that Dish has sub-setted its capabilities.
I think they rely on nothing being final until the commit changes pointers in a single swipe.
(How many pointers? Not that it matters.) Are they really using journaling for changes?

-Ken

 

[Tampa8]

My VIP receivers run better by rebooting every night, my observation after some 13 years or so with Dish. That equipment other people have can run for a year is great, these receivers can not. In fact, even at that, I have all mine set on a timer to do a hard reset sometime after the reboot. Since doing that my receivers are never sluggish anymore, never freeze. So yes, they DO need to be rebooted on some type of regular basis. And contrary to what some are saying, there are many times a change that has been made does not show up until I do a check switch or a reboot. (though sometimes the change does show up)
That the time can't be changed however is the problem for Hopper owners.

 

[jawxx]

I theorize that the non-adjustable reboot has something to do with PrimeTimeAnyTime and the ability to use Auto-Hop for those programs but not until the day after which they were recorded

 

[John Kotches]

dwarren:

Here's a crazy thought. Use the release file. Is the release file the same as yesterday? Don't need to reboot. No need to do this in a complicated fashion.

The one flaw is if they forget to update the release file. In which case you have an issue with procedure. I work in an engineering environment with servers directly facing our customers. Procedure is everything.

Simple and highly effective.

Sent from my Samsung Galaxy Note 2 using Tapatalk 2.x

 

[John Kotches]

Ken,

Please read up on ext3 and ext4fs. I don't know of a single modern file system that doesn't use a journal/transaction log.

It's much faster to replay the journal than it is to do an integrity check.

An fsck wouldn't change the block allocation requirements, it would simply validate that all of the file data blocks are where they are supposed to be.


Sent from my Samsung Galaxy Note 2 using Tapatalk 2.x

 

[John Kotches]

Tampa8:

I'm not saying that they should have an annual reboot cycle, but the fact that they are rebooting daily means something is wrong.

Most people have been brainwashed by years of the Microsoft way of troubleshooting, as if rebooting should be a normal thing. It should be the option of last resort.

Perhaps we should set higher standards for them. I know if my customers were experiencing this I'd be unemployed.

Sent from my Samsung Galaxy Note 2 using Tapatalk 2.x

 

[Tampa8]

I don't totally disagree with what you are saying! My only point, the reality of it is they have to be rebooted, and relatively often, I think being KKlare's point. That my smartphone or laptop does not have to be really has no bearing, and does not mean the receiever does not have to be for whatever the reason is.

 

[neonturbo]

My annoyance is not that it DOES reboot, it is WHEN it reboots. I am quite aware they (for various reasons) reboot every day. My 722 and my previous dish receiver (211K?) before that rebooted every day and I never noticed it for the 10 plus years I have had Dish because I set the time to when I was NOT watching TV.

If you work 2nd shift, this is right in your prime viewing time. Again, that setting in the software doesn't do anything, so why even have it there?

DIRT team or anyone who can get the message to the engineers: Please enable time of day rebooting like the 722 and other receivers.