https://vmk.crm-onebox.com/admin/shop/statistic/
at 14:08 approximately stopped working
I overloaded the hetzner.cloud server, but after a while it worked, delaying orders from sites and payments, and again silence. Overloaded a couple more times On / Off the server and nothing.
CRM does not update data on cron triggering
https://vmk.crm-onebox.com/admin/shop/statistic/ at 14:08 approximately stopped working I overloaded the hetzner.cloud server, but after a while it worked, delaying orders from sites and payments, and again silence. Overloaded a couple more times On / Off the server and nothing. CRM does not update data on cron triggering
Good afternoon. Now cron is working. Kron did not have enough resources to finalize. What exactly led to the shortage of resources is not clear now (perhaps an unload that did not have enough memory). The logic of the system is such that in situations where there was an abnormal termination of the script, the next start attempt will be in 90 minutes. Restarting the server will not affect the operation logic in any way. If the cron worked during the reboot, then the operating system will force it to end, and this will also create a situation, that the next launch will be in 90 minutes.
Good afternoon.
Now cron is working.
Kron did not have enough resources to finalize.
What exactly led to the shortage of resources is not clear now (perhaps an unload that did not have enough memory).
The logic of the system is such that in situations where there was an abnormal termination of the script, the next start attempt will be in 90 minutes.
Restarting the server will not affect the operation logic in any way.
If the cron worked during the reboot, then the operating system will force it to end, and this will also create a situation,
that the next launch will be in 90 minutes.
Tasun Sergey Vladimirovich Employee wrote: Good afternoon. Now cron is working. Kron did not have enough resources to finalize. What exactly led to the shortage of resources is not clear now (perhaps an unload that did not have enough memory). The logic of the system is such that in situations where there was an abnormal termination of the script, the next start attempt will be in 90 minutes. Restarting the server will not affect the operation logic in any way. If the cron worked during the reboot, then the operating system will force it to end, and this will also create a situation, that the next launch will be in 90 minutes.
Is it possible to reduce the period to 10-30 minutes this. 1.5 hours without a working CPM is another problem at the end of the day
[quote]
Tasun Sergey Vladimirovich
Employee wrote:
Good afternoon.
Now cron is working.
Kron did not have enough resources to finalize.
What exactly led to the shortage of resources is not clear now (perhaps an unload that did not have enough memory).
The logic of the system is such that in situations where there was an abnormal termination of the script, the next start attempt will be in 90 minutes.
Restarting the server will not affect the operation logic in any way.
If the cron worked during the reboot, then the operating system will force it to end, and this will also create a situation,
that the next launch will be in 90 minutes.
[/quote]
Is it possible to reduce the period to 10-30 minutes this. 1.5 hours without a working CPM is another problem at the end of the day
Tarovik Dmitry Nikolaevich Client wrote: Is it possible to reduce the period to 10-30 minutes this
Not so simple. In certain situations (uploads, sending a large number of emails, etc.), crons can run for quite a long time. And changing the time can lead to crown doubling, which is also not very good.
[quote]
Tarovik Dmitry Nikolaevich
Client wrote:
Is it possible to reduce the period to 10-30 minutes this
[/quote]
Not so simple.
In certain situations (uploads, sending a large number of emails, etc.), crons can run for quite a long time.
And changing the time can lead to crown doubling, which is also not very good.
Here you need to look at what exactly the script is doing, if these are computational operations or data processing from the database, then yes - 4 threads should give an increase. But keep in mind that working with the database is disk operations, and disk operations are one of the slowest operations in the operating system. If this is work with external services, then it depends more on the speed and quality of the Internet channel. With significant volumes of exchange, changing the echo in the network from 1 ms to 30 ms increases the cron time from 12 minutes to 3 hours. Generally speaking, 4 threads are faster than 2 threads. several services are running in parallel on the server, scripts are running in parallel, etc., then the command queue will be distributed between 4 computing units. How significant the performance gain will be, I'm not ready to answer.
Here you need to look at what exactly the script is doing, if these are computational operations or data processing from the database, then yes - 4 threads should give an increase. But keep in mind that working with the database is disk operations, and disk operations are one of the slowest operations in the operating system.
If this is work with external services, then it depends more on the speed and quality of the Internet channel.
With significant volumes of exchange, changing the echo in the network from 1 ms to 30 ms increases the cron time from 12 minutes to 3 hours.
Generally speaking, 4 threads are faster than 2 threads. several services are running in parallel on the server, scripts are running in parallel, etc., then the command queue will be distributed between 4 computing units.
How significant the performance gain will be, I'm not ready to answer.
Well, the point is clear. if you take the Ukrainian server and the https://www.hetzner.com/ server, then the fact that the CRM is on a server in another country greatly affects the work of the crown?
Well, the point is clear.
if you take the Ukrainian server and the https://www.hetzner.com/ server, then the fact that the CRM is on a server in another country greatly affects the work of the crown?
As stated above, this only affects cron runtime when external integrations are present. If a small number of requests (200) to an external resource are made, then this will not affect the execution time. When executing 40,000 requests, an increase in network echo time can result in a significant increase in execution time.
Tarovik Dmitry Nikolaevich Client wrote: if you take the Ukrainian server and the https://www.hetzner.com/ server, then the fact that the CRM is on a server in another country greatly affects the work of the crown?
The ping time between a server hosted in Ukraine and a server hosted by Hetzner will depend on the number of transitions and the quality of communication channels of providers that provide a route between Ukraine and Germany. This time is about 30 ms to 45 ms. You also need to take into account that Hetzner has 3 data centers and they are located geographically in different places Nuremberg, Falkenstein and Helsinki.
As stated above, this only affects cron runtime when external integrations are present.
If a small number of requests (200) to an external resource are made, then this will not affect the execution time.
When executing 40,000 requests, an increase in network echo time can result in a significant increase in execution time.
[quote]
Tarovik Dmitry Nikolaevich
Client wrote:
if you take the Ukrainian server and the https://www.hetzner.com/ server, then the fact that the CRM is on a server in another country greatly affects the work of the crown?
[/quote]
The ping time between a server hosted in Ukraine and a server hosted by Hetzner will depend on
the number of transitions and the quality of communication channels of providers that provide a route between Ukraine and Germany.
This time is about 30 ms to 45 ms.
You also need to take into account that Hetzner has 3 data centers and they are located geographically in different places
Nuremberg, Falkenstein and Helsinki.
the essence of all questions is how to understand that the system has enough / not enough resources for stable operation, so as not to have force majeure and hang-ups. It hurts work.
the essence of all questions is how to understand that the system has enough / not enough resources for stable operation, so as not to have force majeure and hang-ups. It hurts work.
Tasun Sergey Vladimirovich Employee wrote: View logs /admin/shop/report/logs/ and monitor the server.
Cool, there would be a basic instruction for dummies on how to recognize something from the logs. I am not strong in this direction.
[quote]
Tasun Sergey Vladimirovich
Employee wrote:
View logs /admin/shop/report/logs/
and monitor the server.
[/quote]
Cool, there would be a basic instruction for dummies on how to recognize something from the logs. I am not strong in this direction.
Tarovik Dmitry Nikolaevich Client wrote: By the way, I have not seen such a section in the menu
In the "Reports" menu and here is the section
[quote]
Tarovik Dmitry Nikolaevich
Client wrote:
By the way, I have not seen such a section in the menu
[/quote]
In the "Reports" menu
and here is the section
[file]1656[/file]
Please join the conversation. If you have something to say - please write a comment. You will need a mobile phone and an SMS code for identification to enter.
Log in and comment
Donate
You don't have enough funds in your account Top up