Closed omuravskiy closed 1 year ago
Thanks for the report.
The 2050 limit (number of NLRIs that can easily fit in a 4k BGP update) needs revisited since the extended-messages RFC has been published. It needs a bit of refactoring to move back to dynamic memory allocation in this function.
I'm not exactly sure about the reason for the confusion on the number of prefixes, but I will look at removing the problem first, and then see if it remains afterwards.
@omuravskiy Please could you test PR #8 and in particular, could you check the test_expect/updates.long_withdrawal.gz.bgp.gz
and confirm that this matches what you expected in the long_withdrawal.gz
test input file?
Thanks!
@spakka Yes, the bgpdump of test_data/updates.long_withdrawal.gz
looks correct. Thanks!
This has been fixed in PR #8
Now that the RFC is out it would be nice if bgpdump supports it. Currently when parsing a file longer than 4k it is confused about the number of prefixes in the MP_UNREACH_NLRI attribute (for example), reporting
long_withdrawal.gz