Get the App
SLTechnology News&Howtos  ›  Servers  › 

How to analyze the problems of oradebug diagnosis system

Shulou Source: shulou.com Published: 2022-06-01 02:52:00 11月08日 Update

How to analyze the oradebug diagnostic system problem, for this problem, this article details the corresponding analysis and solution, hoping to help more small partners who want to solve this problem find a simpler and easier way.

oracle hanganalyze

There are several levels of hanganalysis:

10 Dump all processes (IGN state)

5 Level 4 + Dump all processes involved inwait chains (NLEAF state)

4 Level3 + Dump leaf nodes (blockers) in wait chains(LEAF,LEAF_NW,IGN_DMP state)

3 Level2 + Dump only processes thought to be in a hang (IN_HANG state)

1-2 Only HANGANALYZE output, no process dump at all

From the above information, there are many levels to choose when hanganalyzedump, so how to choose the level?

In general, it is not recommended to use operations above level 3, because the resulting trace may be large, especially for large OLTP systems;

In addition, the system pressure may be huge when the database hang lives, so generating a large trace may lead to more serious problems.

systemstate

systemstate dump level profile:

1. Level 2:dump(excluding lock elements)

2. Level 10:dump

3. Level 11:dump+global cache of rac

4. Level 256:short stack

5. Level 258: Level 256+ Level 2

6. Level 266: Level 256+ Level 10

7. Level 267: Level 256+ Level 11

processstate

1. Levels 11 and 267 dump the global cache, producing large trace files and are generally not recommended.

2. In general, if the process is not too many, it is recommended to use 256, because this can dump the function stack of the process, which can be used to analyze what the process is performing, but generating a short stack is time-consuming, if the process is very many, such as 2000 processes, it may take more than 30 minutes. In this case, level 10 or level 258 can be generated, level 258 collects more short stacks than level 10, but collects less lock element data than level 10.

3. Bug 11800959 - A SYSTEMSTATE dump with level >= 10 in RAC dumps huge BUSY GLOBAL CACHE ELEMENTS - can hang/crash instances (Doc ID 11800959.8). This Bug was fixed in 11.2.0.3 for

Tags: Levels questions systems analysis processes data suggestions situations databases large functions stacks times more help generation login solutions choices huge Apple Docker Huawei Linux macOS MariaDB Microsoft MySQL NVidia OPPO Reno Apple Shulou Technology Huawei Docker Shulou Information