Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Fixed
-
10.8(EOL), 10.9(EOL), 10.10(EOL), 10.11, 11.0(EOL), 11.1(EOL), 11.2
Description
This came up while I was preparing to test MDEV-14795 with something like the following:
mkdir /dev/shm/data
|
./mariadbd --datadir /dev/shm/data --innodb-data-file-path=ibdata1:56G --bootstrap < /dev/null
|
The startup message would wrongly claim the following:
10.9 35de8326fb6ca2751c83ee7465c33347b9efeb93 |
2023-07-05 13:39:58 0 [Note] InnoDB: Setting file './ibdata1' size to 0.000GiB. Physically writing the file full; Please wait ...
|
2023-07-05 13:40:19 0 [Note] InnoDB: File './ibdata1' size is now 0.000GiB.
|
This happens because of 32-bit arithmetics in SysTablespace::set_size(). All other messages that use the MDEV-27158 ib::bytes_iec seem to be correct.
Attachments
Issue Links
- is duplicated by
-
MDEV-31627 innodb large file size are have overflowed "size to 0.000B" message in logs
- Closed
- relates to
-
MDEV-27158 Humanize the number of bytes in InnoDB messages
- Closed