Replace master-reboot-down-after-period with primary-reboot-down-after-period in sentinel.conf

[hwware]

Update sentinel.conf config parameter,

From:

SENTINEL master-reboot-down-after-period mymaster 0

To:

SENTINEL primary-reboot-down-after-period myprimary 0

But we still keep the backward compatibility, clients could use SENTINEL master-reboot-down-after-period mymaster 0 OR
SENTINEL primary-reboot-down-after-period myprimary 0

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 8 lines in your changes missing coverage. Please review.

Project coverage is 70.27%. Comparing base (dd4bd50) to head (8cb77fe).

Additional details and impacted files

@@             Coverage Diff              @@
##           unstable     #647      +/-   ##
============================================
+ Coverage     70.11%   70.27%   +0.16%     
============================================
  Files           112      112              
  Lines         60587    60590       +3     
============================================
+ Hits          42480    42581     +101     
+ Misses        18107    18009      -98     

Files	Coverage Δ
src/sentinel.c	0.00% <0.00%> (ø)

... and 9 files with indirect coverage changes

[zuiderkwast]

LGTM.

There's still no primary alias for GET-MASTER-ADDR-BY-NAME? (#36)

There's some duplicated code in the test case. We could use a vaiable for primary/master, like this, if you want:

foreach role {primary master} {
    # ...
    S $id SENTINEL SET mymaster $role-reboot-down-after-period 5000
    # ...
}

[zuiderkwast]

3b2197d doesn't look right. Before this commit, the test case repeated the whole check for "master" and "primary". With the for loop it only repeats 3 commands for "primary" and "master".

I think we need to put this foreach role {primary master} around more code, until the end of the test case, so we actually test everything for primary and everything again for master. Makes sense?

    foreach role {master primary} {
        foreach_sentinel_id id {
            S $id SENTINEL SET mymaster $role-reboot-down-after-period 5000
            ...
        }

        kill_instance valkey $master_id
        reboot_instance valkey $master_id
    
        foreach_sentinel_id id {
            ...
        }

        ...

        # Make sure the instance load all the dataset
        while 1 {
            ...
        }
    }
} ; # <--- end of test case

[javsalgar]

Hi! Trying this on version 8.0, I see that it crashes with the default sentinel.conf values

*** FATAL CONFIG FILE ERROR (Version 8.0.0) ***
Reading the configuration file, at line 358
>>> 'SENTINEL primary-reboot-down-after-period myprimary 0'
No such master with specified name.

To fix I need to replace myprimary with mymaster

[enjoy-binbin]

@javsalgar thanks for the report. I will take a look later