Bug #73783 ndb_redo_log_reader documentation is not correct

• Submitted: 1 Sep 2014 14:03
• Modified: 10 Jan 2015 13:05
• Reporter: Daniël van Eeden (OCA)
• Status: Duplicate
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S2 (Serious)
• Version: 7.2
• OS: Any
• Assigned to: Jon Stephens
• CPU Architecture: Any

[1 Sep 2014 14:03] Daniël van Eeden

Description:
Page: http://dev.mysql.com/doc/refman/5.5/en/mysql-cluster-programs-ndbd-redo-log-reader.html

1. The command is named ndb_redo_log_reader instead of ndbd_redo_log_reader
2. The command has more options than listed in the documentation
3. The documenation lists "ndb_#_fs/D#/LCP/#/T#F#.Data"
But it probably should be "ndb_#_fs/LCP/#/T#F#.Data"

How to repeat:
ndb_redo_log_reader

Usage:
ndb_redo_log_reader <Binary log file> [-noprint] [-nocheck] [-mbyte <0-15>] [-mbyteheaders] [-pageheaders] [-filedescriptors] [-page <0-31>] [-pageindex <12-8191>] -twiddle

Error in redoLogReader(). Exiting!

[2 Sep 2014 6:27] MySQL Verification Team

Hello Daniël,

Thank you for the report.
Verified as described.

Thanks,
Umesh

[2 Sep 2014 6:29] MySQL Verification Team

// ndb_redo_log_reader

mysql-cluster-7.3.5]$ ls -l bin/ndb_r*
-rwxr-xr-x 1 ushastry ushastry  4862652 Apr  1 21:47 bin/ndb_redo_log_reader

// command-line options for the ndb_redo_log_reader program

mysql-cluster-7.3.5]$ bin/ndb_redo_log_reader --help

This command reads a redo log file, checking it for errors, printing its contents in a human-readable format, or both.
Usage: ndbd_read_log_reader [OPTIONS]:
bin/ndb_redo_log_reader <Binary log file> [-noprint] [-dump] [-twiddle] [-lap] [-nocheck] [--help] [-mbyte <0-15>] [-mbyteheaders] [-pageheaders] [-filedescriptors] [-page <0-31>]  [-pageindex <12-8191>]

// From local repo

storage]$ sudo tree  ndb_10_fs/LCP
ndb_10_fs/LCP
+-- 0
¦   +-- T10F0.Data
¦   +-- T10F1.Data
¦   +-- T14F0.Data
¦   +-- T14F1.Data
¦   +-- T20F0.Data
¦   +-- T20F1.Data
¦   +-- T2F0.Data
¦   +-- T2F1.Data
¦   +-- T3F0.Data
¦   +-- T3F1.Data
¦   +-- T4F0.Data
¦   +-- T4F1.Data
¦   +-- T5F0.Data
¦   +-- T5F1.Data
¦   +-- T7F0.Data
¦   +-- T7F1.Data
¦   +-- T8F0.Data
¦   +-- T8F1.Data
¦   +-- T9F0.Data
¦   +-- T9F1.Data
+-- 1
    +-- T10F0.Data
    +-- T10F1.Data
    +-- T14F0.Data
    +-- T14F1.Data
    +-- T20F0.Data
    +-- T20F1.Data
    +-- T2F0.Data
    +-- T2F1.Data
    +-- T3F0.Data
    +-- T3F1.Data
    +-- T4F0.Data
    +-- T4F1.Data
    +-- T5F0.Data
    +-- T5F1.Data
    +-- T7F0.Data
    +-- T7F1.Data
    +-- T8F0.Data
    +-- T8F1.Data
    +-- T9F0.Data
    +-- T9F1.Data

2 directories, 40 files

[4 Sep 2014 11:20] Jon Stephens

This is in fact a software bug. The correct name of the program is ndbd_redo_log_reader. The reason for this is that programs whose names are prefixed with "ndb_" are intended to be used with a running cluster, whereas this program does not, and it is prefixed with "ndbd_" for this reason and because it reads a REDO log produced by ndbd. 

The information I have given above was provided to me in person (and very nearly word for word) by Jonas Oreland ca. 2006. 

The documentation is correct: an error was introduced into the source so that this program was built with the wrong name. Please fix the program as per the bug I submitted against this very defect in December. Thanks.

[4 Sep 2014 11:26] Daniël van Eeden

What is the bug number for the bug which causes the tool to be build with the wrong name?

[10 Jan 2015 10:33] Daniël van Eeden

From which bug is this a duplicate?

[10 Jan 2015 13:05] MySQL Verification Team

This is duplicate of internally reported Bug 17921572