[Date Prev][Date Next] [Chronological] [Thread] [Top]

AW: bdb bad performance again (really need help)



Hallo Quanah,

It looks like a bug in berkeley. I wonder if it has been reported to them? Horward wrote once that there is a number of bugs in BDB which could cause observed behaviour of the backend, however is anybody sure that the problem we observe is caused by the known bugs? 

I understood Horwards message as an instruction to

1) increase cache size
2) wait for Berkeley DB 4.2

And what if Berkeley will still have the same bug in 4.2? I would report this problem to them myself, however I don't even know correct words to use when describing the problem, just very few rather "zoological" observations (like it does not work when it rains) .....

Best regards, vadim tarassov.

-----Ursprüngliche Nachricht-----
Von: Quanah Gibson-Mount [mailto:quanah@stanford.edu]
Gesendet: Montag, 10. November 2003 08:42
An: Stefan Metzmacher; openldap-software@OpenLDAP.org
Betreff: Re: bdb bad performance again (really need help)




--On Monday, November 10, 2003 8:10 AM +0100 Stefan Metzmacher 
<metze@metzemix.de> wrote:

>> We do not put our entire DB into memory, however, we do run a very large
>> cache (2.5GB).
>>
>> The reason is essentially this -- If you do not have a sufficient cache
>> size, the system must read almost continuously from disk to answer
>> queries. This trashes the disk, and eventually has the end effect of
>> corrupting the database.
>
> How can read operations corrupt the data base?

I don't explain it, I just know what has happened for us, and plenty of 
other sites as well.  I think it is more that BDB's disk cache gets 
corrupted, more than the underlying database files.

--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/TSS/Computing Systems
ITSS/TSS/Infrastructure Operations
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html