How to rebuild the error and recovery database

[Rick Brown]

Site Cleanup with rebuild procedures

1.   Type showroot to make sure root and site are set correctly

2.   Stop all processes – hcienginestop -p (for all processes)

3.   Run hciprocstatus (to verify all engines are down)

4.   Stop Daemons – hcisitectl -K

5.   Run hcisitectl (to make sure lock manager and monitor daemon are down)

6.   Clear shared memory region – hcimsiutil -R

7.   Change to the database directory – /hci/root3.7.XP//exec/databases

8.   Remove vista.taf file – rm vista.taf

9.   Change directory to exec directory – cd ..

10.  Remove the monitorShmemFile – rm monitorShmemFile

11.  Reinitialize the ICL Database – hcidbinit

Topic

[James Nelson]

do the ‘keybuild/dchain’ combinations have the same effect as an ‘hcidbinit’ command without destoying the data?

[Rick Brown]

James,

Basically, yes.

The hcidbinit -AC command generates a new set of databases and you loose all data that was in them.

The rebuild commands will reindex the databases in an attempt to save what is in the databases.  Always try this before you write them off as corrupt.

[James Nelson]

Thanks.

Actually, we don’t do the ‘hcidbinit -AC’ command because of corruption so much as to compact the database.  We do this as part of our monthly maintenance.

[Anonymous]

I was wondering how one could cleanup large rlog* and elog* files in databases directory. At our place they grow to more 30 MB while parsing large XML files. I would appreciate any help.

Thanks

./dinakar

[Author]

Replies