Shulou(Shulou.com)06/01 Report--

Preface

For Oracle clusters, the explanation of diagwait can be easily found on the Internet:

This problem occurs only in versions prior to ORACLE 11.2, and in 11G R2, the value of diagwait is configured to 13 by default.

For versions prior to 11.2, diagwait needs to be manually modified to 13 to postpone the restart to allow enough time to write the log information in the cache to disk files, and to reduce the possibility of restart due to the short allowed time to interact with OS.

Diagwait is a small knowledge point, but it is universal. Because many customers' production database version is lower than 11.2, and the diagwait is not set correctly (the default configuration is unreasonable), some cases of node expulsion lack relevant logs and cannot accurately analyze the reasons, or the default 1.5s (1000ms+500ms) of oporcd is too short and too sensitive, for example, excessive instantaneous pressure can easily lead to node host restart, which is obviously not what we expected.

Diagwait setting method (using root users):

Crsctl set css diagwait 13

Diagwait query method:

Crsctl get css diagwait ps-ef | grep oproc

The cluster needs to be restarted to take effect.

The following process is verified in the test environment:

[oracle@rac1-server ~] $ps-ef | grep oprocoracle 12527 6205 0 11:44 pts/4 00:00:00 grep oprocroot 21426 20189 0 Jan24? 00:00:00 / bin/sh / etc/init.d/init.cssd oprocdroot 21799 21426 0 Jan24? 00:00:15 / s01/oracle/product/10.2.0/crs_1/bin/oprocd.bin run-t 1000-m 500-f [oracle@rac1-server ~] $crsctl get css diagwaitConfiguration parameter diagwait is not defined. [root@rac1-server ~] # / s01/oracle/product/10.2.0/crs_1/bin/crsctl set css diagwait 13Configuration parameter diagwait is now set to 13. [root@rac1-server ~] # / s01/oracle/product/10.2.0/crs_1/bin/crsctl get css diagwait 13 [oracle@rac1-server ~] $ps-ef | grep oprocoracle 15375 14082 0 12:56 pts/4 00:00:00 grep oprocroot 21426 20189 0 Jan24? 00:00:00 / bin/sh / etc/init.d/ Init.cssd oprocdroot 21799 21426 0 Jan24? 00:00:15 / s01/oracle/product/10.2.0/crs_1/bin/oprocd.bin run-t 1000-m 500-f you can see that the setting of oprocd has not been modified. -- crs needs to be restarted (all nodes need to be restarted): [root@rac1-server ~] # / s01/oracle/product/10.2.0/crs_1/bin/crsctl stop crs [root@rac1-server ~] # / s01/oracle/product/10.2.0/crs_1/bin/crsctl start crs-- check the oprocd process again and see that the value after-m has changed from the default 500 to 10000. Unit is ms: [oracle@rac1-server ~] $ps-ef | grep oprocdroot 17520 16659 0 12:58? 00:00:00 / bin/sh / etc/init.d/init.cssd oprocdroot 17916 17520 0 12:58? 00:00:00 / s01/oracle/product/10.2.0/crs_1/bin/oprocd.bin run-t 1000-m 10000-hsi 5 grep oprocd 10 ps 50 ps 90-foracle 18504 16764 0 12:59 pts/4 00:00:00 grep oprocd

Summary

The above is the whole content of this article, I hope that the content of this article has a certain reference and learning value for your study or work, if you have any questions, you can leave a message and exchange, thank you for your support.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.