Jack, Alex and Jan:

Should we make the Sparc boot archive dcfs bug had a dependency on 13109 ???

Then we know when 13109 is fixed and Alex can fix the sparc dcfs bug.


On 02/17/10 03:06 PM, Jack Schwartz wrote:
> Hi Alex and Jan.
>
> The DDU uses pkg(5), and right now pkg(5) requires some extra space 
> for its catalog caching operations.  Bug 13109 documents this to be 
> ~90Mb extra ramdisk space now, which is too large to work.  However, 
> even with a fix some extra ramdisk space will likely be needed.  Since 
> no fix is implemented yet I can't say how much.  Unless it is urgent, 
> I suggest holding off on changing the boot archive size until 13109 
> has been fixed or at least better addressed.
>
>     I hope that helps,
>     Jack
>
>
>
> On 02/17/10 02:29, Alexander Eremin wrote:
>> On Wed, 2010-02-17 at 10:57 +0100, Jan Damborsky wrote:
>>> Hi Alex,
>>>
>>> looking at this bug and fixed related issues, I agree we could now
>>> get rid of dcfs assuming that we can sacrifice free space in ramdisk
>>> we are going to lose (~ 25MB):
>>>
>>> [1] We need to make sure that there are no issues with
>>>       respect to 'supposed to work' configurations.
>>>
>>> Currently, installers (AI/text) are supposed to work on Sparc
>>> machines with 512MB of memory.
>>>
>>> I don't assume that fix for 6361 itself would cause problems,
>>> since it does not change the size of ramdisk, but we should check
>>> that we can still install on those machines with latest bits
>>> in order to confirm we can afford losing those ~25MB which
>>> (if needed) could be used to decrease size of ramdisk.
>>>
>>> I think Mary has access to machines which we could use for testing -
>>> we could help you with this.
>>>
>>> [2] We should check with DDU project which could introduce potential
>>>       consumers of ramdisk space.
>>>
>>> Jack, do you happen to know how DDU might affect size boot archive
>>> and free space in ramdisk ?
>>>
>>> Thank you,
>>> Jan
>> Thanks Jan,
>> agree, so waiting answer from Jack and it's would be fine if
>> Mary could test boot_archive without dcfs on 512MB.
>>
>>
>

Reply via email to