V4REV 4.7 - Crash when DB reaches size of 2147483647
Ruslan Zasukhin
ruslan_zasukhin at valentina-db.com
Thu Nov 21 01:32:01 CST 2013
On 11/21/13, 3:15 AM, "Bernard Devlin" <bdrunrev at gmail.com> wrote:
Hi Bernard,
1) so it sounds that you can give us something that reproduce this crash...
7 iteration adding 90K records
2) what worries is v4.7 ... Quite old (2-3 years?) engine
First of all it needs check this iterations against latest 5.4.3
Possible?
> I'm checking if Valentina will provide the required insert & select speed for
> a database which could grow into the 50gb range.
>
> The database contains Long Long PK fields, then the rest of the fields are
> Varchar and Text fields. The only indexes are those required for PK
> uniqueness.
>
> During a bulk insert, I consistently get a crash when the database reaches a
> size of 2147483647 <tel:2147483647> . I note this is the maximum size of a
> Long.
>
> The database has a segment size of 32768. And I get the crash regardless of
> whether the database consists of 1 file (kDscDatBlbInd) or 4 files
> (kDsc_Dat_Blb_Ind).
>
> The table with the most records in it has 30801659 records, and is 2.00gb in
> size.
>
> There are about 90,000 distinct records, and for this test I'm inserting the
> same set of records over and over again. On about the 7th iteration of
> inserting this 90,000 records, the crash occurs. Thus, it seems it is nothing
> in any of the individual inserts which causes the crash (or it would crash
> during the earlier iterations)
>
> That this db size is exactly that of a Long, makes me think I'm hitting some
> magic limit. My understanding is that a 32k segment size should be able to
> produce a db that approximates to 64gb of data.
>
> The database is encrypted. The database is on a NTFS partition.
--
Best regards,
Ruslan Zasukhin
VP Engineering and New Technology
Paradigma Software, Inc
Valentina - Joining Worlds of Information
http://www.paradigmasoft.com
[I feel the need: the need for speed]
More information about the Valentina
mailing list