[gentoo-catalyst] [PATCH] lock: fix hardlink_active definition

[gentoo-catalyst] [PATCH] lock: fix hardlink_active definition

[Mike Frysinger]

This is a member function of a class, so it must have a |self| argument
first.  The current code is clearly broken otherwise.  Not that it seems
to matter as there aren't any callers in the current codebase.
---
 catalyst/lock.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/catalyst/lock.py b/catalyst/lock.py
index a6f9d3a..9f954eb 100644
--- a/catalyst/lock.py
+++ b/catalyst/lock.py
@@ -354,7 +354,7 @@ class LockDir:
 				return True
 		return False
 
-	def hardlink_active(lock):
+	def hardlink_active(self, lock):
 		if not os.path.exists(lock):
 			return False
 
-- 
2.5.2

Re: [gentoo-catalyst] [PATCH] lock: fix hardlink_active definition

[Brian Dolbec]

On Mon,  5 Oct 2015 23:48:07 -0400
Mike Frysinger <vapier@gentoo.org> wrote:

> This is a member function of a class, so it must have a |self|
> argument first.  The current code is clearly broken otherwise.  Not
> that it seems to matter as there aren't any callers in the current
> codebase. ---
>  catalyst/lock.py | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/catalyst/lock.py b/catalyst/lock.py
> index a6f9d3a..9f954eb 100644
> --- a/catalyst/lock.py
> +++ b/catalyst/lock.py
> @@ -354,7 +354,7 @@ class LockDir:
>  				return True
>  		return False
>  
> -	def hardlink_active(lock):
> +	def hardlink_active(self, lock):
>  		if not os.path.exists(lock):
>  			return False
>
  
:)

If you want to re-do all the lock code, go for it.  I know Anthony was
looking at putting that on to his TODO list.  Maybe the two of you can
get it all in great shape.

-- 
Brian Dolbec <dolsen>

Re: [gentoo-catalyst] [PATCH] lock: fix hardlink_active definition

[Anthony G. Basile]

On 10/6/15 2:41 AM, Brian Dolbec wrote:
> On Mon,  5 Oct 2015 23:48:07 -0400
> Mike Frysinger <vapier@gentoo.org> wrote:
>
>> This is a member function of a class, so it must have a |self|
>> argument first.  The current code is clearly broken otherwise.  Not
>> that it seems to matter as there aren't any callers in the current
>> codebase. ---
>>   catalyst/lock.py | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/catalyst/lock.py b/catalyst/lock.py
>> index a6f9d3a..9f954eb 100644
>> --- a/catalyst/lock.py
>> +++ b/catalyst/lock.py
>> @@ -354,7 +354,7 @@ class LockDir:
>>   				return True
>>   		return False
>>   
>> -	def hardlink_active(lock):
>> +	def hardlink_active(self, lock):
>>   		if not os.path.exists(lock):
>>   			return False
>>
>    
> :)
>
> If you want to re-do all the lock code, go for it.  I know Anthony was
> looking at putting that on to his TODO list.  Maybe the two of you can
> get it all in great shape.
>
we should write a separate class for file locking with a timeout. the 
motivation here is bug #519656.  I wrote a sketch of what the class 
would do, but I have been more involve with working on GRS and just 
haven't gotten to it.  also on the TODO list is to get catalyst using 
cgroup.  catalyst just doesn't die well on TERM or INT.  with GRS I use 
cgroups and its amazingly clean.  one of the best features added to the 
kernel in recent history.

-- 
Anthony G. Basile, Ph.D.
Gentoo Linux Developer [Hardened]
E-Mail    : blueness@gentoo.org
GnuPG FP  : 1FED FAD9 D82C 52A5 3BAB  DC79 9384 FA6E F52D 4BBA
GnuPG ID  : F52D4BBA