Hello, I was chasing other IPVS issues while preparing to finalize these changes. I plan to do tests this weekend and to submit my patch but without gso_size modifications. Can post patch for this ch
Hi Julian, Do you have a chance to work on the IPv6 side? Thanks, Martin -- To unsubscribe from this list: send the line "unsubscribe lvs-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx
Hello, if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU)) ort = ort->from; Sorry, my fault, I missed above re-assignment... WARN_ON_ONCE(ort->rt6i_flags & (RTF_CACHE | RTF_PCPU)); Aha, ok. I thought, only
hmm...I am not sure I follow this bits. Where is the warn? Note that "nrt6" and "from" are passed to rt6_insert_exception() instead of "rt6". The existing rt6_do_update_pmtu() looks correct. The mtu
Hello, I checked again the code and it looks like sockets are forced to use new exceptional route (RTF_CACHE/fnhe) via dst_check only when the PMTU update should move them away from old non-exception
I agree; I think it is fine. A route is route. The IPVS use cases with local redirects are blurring the line with needs for local routes. -- To unsubscribe from this list: send the line "unsubscribe
IPv6 used not to create cache route for DST_HOST route which is a /128 route (that includes local /128 route). Because of this, it had a bug such that a PMTU for the DST_HOST route will trigger dst.o
Hello, Probably. I completely forgot the IPv6 part but as I don't know the IPv6 code enough, it may take some time to understand what can be the problem there... I'm not sure whether everything start
For IPv6, the 'if (rt6->rt6i_flags & RTF_LOCAL)' gate in __ip6_rt_update_pmtu() may need to be lifted also. -- To unsubscribe from this list: send the line "unsubscribe lvs-devel" in the body of a me
Hello, In the last week I analyzed the situation and found that just changes in route.c are able to solve the problems, at 99% :) I'm posting a separate patch for this. Here is what happens, I'm test