[cfarm-users] gcc104: rosetta and disk space concerns

[Jacob Bachmeyer]

Zach van Rijn via cfarm-users wrote:
> On Mon, 2022-10-24 at 17:29 +0200, Pierre Muller wrote:
>   
>> Le 24/10/2022 à 17:04, Zach van Rijn a écrit :
>>     
>>> ...
>>>       
>>> I hope that your tests are not interrupted when I clear the
>>> cache. I'll do this once every day or two until a better
>>> solution can be found.
>>>
>>> ...
>>>       
>>   It might be helpful if you could send me/others a
>> notification at the time at which you do the cache cleaning so
>> that we can report back to you if it has a direct impact or
>> not.
>>     

I do not currently run tests on M1 hardware, although I am still 
thankful for what is clearly a useful resource to the community.

> I've just cleared the cache:
>
> gcc104:~ root# ./rosetta_check 
> 131G    /System/Volumes/Data/private/var/db/oah/
> gcc104:~ root# ./rosetta_clear 
> gcc104:~ root# ./rosetta_check 
> 4.0K    /System/Volumes/Data/private/var/db/oah/
>   

Are those scripts simple enough that you could post them here?  I might 
have some ideas that could be helpful, or I might be able to write a 
quick cleanup tool that will purge any cache entries not used recently.  
If the tests are as I suspect (many binaries built/run once and then 
discarded) then I ask how Rosetta knows when to use a cached artifact 
and could we thereby link the cached artifacts back to their original 
binaries and then quickly remove artifacts that derive from binaries 
that are no longer on the system?  (An hourly cron job ought to serve 
nicely here.)  Purging pre-translated "shadow" copies of erased binaries 
also sounds like something Rosetta should do on its own; have you tried 
complaining to Apple about this?

Pierre, can you give more information on the problem at hand?  What 
types of tests are you running and why do they produce so many artifacts 
in the cache?


-- Jacob