matrixone icon indicating copy to clipboard operation
matrixone copied to clipboard

[Feature Request]: Support database actions log wrting

Open domingozhang opened this issue 3 years ago • 3 comments

Is there an existing issue for the same feature request?

  • [X] I have checked the existing issues.

Is your feature request related to a problem?

No response

Describe the feature you'd like

As current design, all the database action can be traced into tables. As an extended requirement, this action also needs to be written into a log file.

  • The log file has to record import actions of MO and critical internal error with account name, event type, timestamp and what happens.
  • All the important actions of MO need to be recorded, including all DDL, all user operations, all nodes operations, all parameter modifications.
  • Critical internal errors including the event disturbs MO working normally, such as removing a datafile manually or one node down.

Describe implementation you've considered

No response

Documentation, Adoption, Use Case, Migration Strategy

No response

Additional information

No response

domingozhang avatar Jul 19 '22 12:07 domingozhang

  • [x] The log file has to record import actions of MO and critical internal error with account name, event type, timestamp and what happens.
    • [x] error
    • [x] with account
    • [x] event type
    • [x] timestamp
    • [x] what happen
  • [x] All the important actions of MO need to be recorded, including all DDL, all user operations, all nodes operations, all parameter modifications.
  • [x] Critical internal errors including the event disturbs MO working normally, such as removing a datafile manually or one node down.
    • [x] MO working normally
    • [x] removing a datafile manually
    • [x] node down

xzxiong avatar Sep 19 '22 06:09 xzxiong

HAkeeper can output MO working status and which nodes are down. But these are not printed currently, I will add this.

w-zr avatar Oct 13 '22 09:10 w-zr

  • [x] The log file has to record import actions of MO and critical internal error with account name, event type, timestamp and what happens.

    • [x] error
    • [x] with account
    • [x] event type
    • [x] timestamp
    • [x] what happen
  • [x] All the important actions of MO need to be recorded, including all DDL, all user operations, all nodes operations, all parameter modifications.

  • [ ] Critical internal errors including the event disturbs MO working normally, such as removing a datafile manually or one node down.

    • [x] MO working normally
    • [ ] removing a datafile manually
    • [x] node down

"removing a datafile manually" those file exception, fileservice will log and return not found message. @LeftHandCold could improve those mesage with filepath or db.table info in following version.

xzxiong avatar Oct 13 '22 09:10 xzxiong