You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@skywalking.apache.org by GitBox <gi...@apache.org> on 2021/03/16 09:31:35 UTC

[GitHub] [skywalking] OPSTime opened a new issue #6557: A requirement of "slow statements"

OPSTime opened a new issue #6557:
URL: https://github.com/apache/skywalking/issues/6557


   
   ### Requirement or improvement
   1、Whether we can add a display of SQL execution times in "slow statements" or sort according to the execution times, in practical work, we pay more attention to those slow queries that are executed the most times. Some slow queries may be executed once a day or at a longer interval, and the relative priority may not be the highest.
   2、Whether "slow statements" can be associated with specific services (for example, add a "current service" to the right of "current database" as an optional condition), so that when troubleshooting problems, we can quickly locate which service slow SQL comes from.
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] OPSTime commented on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
OPSTime commented on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800183220


   1、获取采样数据中该慢SQL出现的次数就能反应出其调用频次高低了，也是有很强参考价值的
   2、和具体服务关联起来，方便快速定位，不能说看到慢SQL再在大脑里回忆或者去查代码去看是和哪个有关系吧，还有一个库里有多种多张表，不同表对应不同服务在使用，不可能完全做到一个服务对应一个数据库实例吧。


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] wu-sheng closed issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
wu-sheng closed issue #6557:
URL: https://github.com/apache/skywalking/issues/6557


   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] wu-sheng commented on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
wu-sheng commented on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800300704


   Please keep the discussion in English. We need things open for the global.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] OPSTime edited a comment on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
OPSTime edited a comment on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800183220


   1、获取采样数据中该慢SQL出现的次数就能反应出其调用频次高低了，也是有很强参考价值的
   2、和具体服务关联起来，方便快速定位，不能说看到慢SQL再在大脑里回忆或者去查代码去看是和哪个有关系吧，还有一个库里有多种多张表，不同表对应不同服务在使用，也可能是在同一实例的不同库中，不可能完全做到一个服务对应一个数据库实例吧，实际生产环境中要考虑业务运行成本。


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] OPSTime removed a comment on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
OPSTime removed a comment on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800183220


   1、获取采样数据中该慢SQL出现的次数就能反应出其调用频次高低了，也是有很强参考价值的
   2、和具体服务关联起来，方便快速定位，不能说看到慢SQL再在大脑里回忆或者去查代码去看是和哪个有关系吧，还有一个库里有多种多张表，不同表对应不同服务在使用，也可能是在同一实例的不同库中，不可能完全做到一个服务对应一个数据库实例吧，实际生产环境中要考虑业务运行成本。


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] wu-sheng commented on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
wu-sheng commented on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800117170


   1. This is not easy to do. Because the slow statement is based on a sampling mechanism. It is hard to get the execution time.
   2. You have to change codes to do that. You could set up the top N metrics to get the SQL list on the service level dashboard. The reason we don't do this, because accessing the same database from different services is not recommended in the distributed architecture. I am not sure why you are doing this.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org

[GitHub] [skywalking] OPSTime commented on issue #6557: A requirement of "slow statements"

Posted by GitBox <gi...@apache.org>.
OPSTime commented on issue #6557:
URL: https://github.com/apache/skywalking/issues/6557#issuecomment-800756004


   1. The frequency of slow SQL call can be reflected by the frequency of slow SQL in sampling data, which is also of great reference value
   
   2. After being associated with specific services, it is convenient to quickly locate problems. Do we have to recall in our brain or look up the code to find out which service is related to? There are also multiple tables in a database. Different tables correspond to different services in use, or in different databases of the same instance. It is impossible to completely realize that one service corresponds to one database instance. Cost should be considered in the production environment.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org