[Gluster-users] volume start causes glusterd to core dump in 3.5.0

[Kaushal M]

Just tested this. It is the use of MD5 in FIPS mode which is causing the
crash. I'll open a bug so that this can be tracked.

~kaushal


On Wed, Apr 30, 2014 at 10:01 AM, Kaushal M <kshlmster at gmail.com> wrote:

> I know about FIPS only by name and I'm not familiar with it.
>
> A simple google search reveals that MD5 is not FIPS compliant and cannot
> be used in FIPS enabled mode. Can someone confirm this?
>
> ~kaushal
>
>
> On Wed, Apr 30, 2014 at 5:25 AM, Justin Clift <justin at gluster.org> wrote:
>
>> On 29/04/2014, at 10:54 PM, Matthew Rinella wrote:
>> > I have this result:
>> >
>> > openssl-1.0.1e-16.el6_5.7.x86_64
>>
>> Hmmm, that's the same version as a host here that's working.
>>
>>
>> > So Im going to admit I also have FIPS mode enabled on my hosts, which
>> is necessary, unfortunately.  This causes a huge problem with puppet
>> because of the crypto algorithms allowed/disallowed.   Im wondering if this
>> is going to be an issue with gluster as well.
>>
>> Personally, I'm not sure.  This *might* be the case atm.
>>
>> This sounds like something the product guys (Red Hat Storage) would
>> be interested in, since playing well with FIPS mode may be needed
>> for some financial institution/government type customers.
>>
>> Niels/Kaushal, any ideas?
>>
>> + Justin
>>
>> --
>> Open Source and Standards @ Red Hat
>>
>> twitter.com/realjustinclift
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20140430/a63be785/attachment.html>