- Регистрация
- 04.11.2010
- Сообщения
- 2 382
- Благодарностей
- 915
- Баллы
- 113
Should be supported!
Many of us are familiar to the problem of suspending instances.
When the instance suspends on any point, the green circle glows, and then nothing happens for a very long time, or never.
- There are a huge amount of resources, and no matter on how many threads ZP is working.
- There is white screen presented, the resource that being manipulated or proxy error.
There were some proposals on the forum:
- Set the number of threads in one process equal to 1 (Done, to the prejudice of minimizing the consumption of the system resources)
- Restart instances more often.—«<The minimum number of tasks to restart equal to 1, Completed tasks before the force reboot qual to 1»>--
So, let it also be done, but I think that many people that have powerful PC's which run ZP know that the setting of frequent rebooting instances,
their total number is not rising above 40 threads (personal experience: the number varies depending on the complexity of the pattern, ie, the duration of its processing)
With my current pattern it's 15-25 threads, although the capacity allows it to run more than 100.
So the very frequent rebooting does not help and is not an option in general.
Works through a proxy, and there is no check without proxy - bans immidiately.
Some instances are likely to stop working after an indefinite time, but suspend from a few minutes (and it's very long) to tens of minutes.
My suggestion is the following - to enter the opportunity of 2 runners in ZP in the settings of instance directly, or set the number of seconds manually:
- To restart the instance forcibly after XXX operation time, ie, for example, if the instance worked for 5 minutes without a reboot, then restart it forcibly,
despite the stoppage of the template, without waiting for its completion.
- To restart the instance forcibly, if from the moment of execution of the recent action about XXX minutes passed.
I.e. if in the template some action was executed, and that the following did not happen in the next 3 minutes, then restart the instance.
It is natural to make these settings optional. Well, that is, for example, if 0 set, then the work of instances is not paying attention to these limits.
The spread of time limits in the settings should be a very big one, from a few seconds to tens of minutes, as all the patterns have very different purposes.
P.S. I hope I've explained everything clearly.
P.P.S. I was looking forward, gritting my teeth, to the fix in new update, but the suspension of instances remained.
I urgely ask for the VERY HOT FIX, only with the introduction of this functional.
It's very necessary!!!
I'm sure that such problems with threads many of us have, so I urge all to response,
for the administration made a decesion to introduce this functionality in the shortest possible time.
Many of us are familiar to the problem of suspending instances.
When the instance suspends on any point, the green circle glows, and then nothing happens for a very long time, or never.
- There are a huge amount of resources, and no matter on how many threads ZP is working.
- There is white screen presented, the resource that being manipulated or proxy error.
There were some proposals on the forum:
- Set the number of threads in one process equal to 1 (Done, to the prejudice of minimizing the consumption of the system resources)
- Restart instances more often.—«<The minimum number of tasks to restart equal to 1, Completed tasks before the force reboot qual to 1»>--
So, let it also be done, but I think that many people that have powerful PC's which run ZP know that the setting of frequent rebooting instances,
their total number is not rising above 40 threads (personal experience: the number varies depending on the complexity of the pattern, ie, the duration of its processing)
With my current pattern it's 15-25 threads, although the capacity allows it to run more than 100.
So the very frequent rebooting does not help and is not an option in general.
Works through a proxy, and there is no check without proxy - bans immidiately.
Some instances are likely to stop working after an indefinite time, but suspend from a few minutes (and it's very long) to tens of minutes.
My suggestion is the following - to enter the opportunity of 2 runners in ZP in the settings of instance directly, or set the number of seconds manually:
- To restart the instance forcibly after XXX operation time, ie, for example, if the instance worked for 5 minutes without a reboot, then restart it forcibly,
despite the stoppage of the template, without waiting for its completion.
- To restart the instance forcibly, if from the moment of execution of the recent action about XXX minutes passed.
I.e. if in the template some action was executed, and that the following did not happen in the next 3 minutes, then restart the instance.
It is natural to make these settings optional. Well, that is, for example, if 0 set, then the work of instances is not paying attention to these limits.
The spread of time limits in the settings should be a very big one, from a few seconds to tens of minutes, as all the patterns have very different purposes.
P.S. I hope I've explained everything clearly.
P.P.S. I was looking forward, gritting my teeth, to the fix in new update, but the suspension of instances remained.
I urgely ask for the VERY HOT FIX, only with the introduction of this functional.
It's very necessary!!!
I'm sure that such problems with threads many of us have, so I urge all to response,
for the administration made a decesion to introduce this functionality in the shortest possible time.
