release v1.9.6 #933
Merged
release v1.9.6 #933
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…T_DESTS, we had set outlen after hit the vs
…mpiling, and fix some bugs Signed-off-by: ywc689 <[email protected]>
add/update agent about socket message
Signed-off-by: ywc689 <[email protected]>
The config file block "ipc_msg/unix_domain" was replaced with command option "-p, --pid-file FILE" in commit 6ce4dd9. Signed-off-by: ywc689 <[email protected]>
It takes long for some nics to update multicast ether addresses. The per-cpu process of per-cpu inetaddr causes mc ether address updating from both slave lcores and master lcore, which is nonsense but a dramatic hurt for performacne. So just do it once on master lcore. Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
It has been noted that rte_eth_dev_set_mc_addr_list is of some restrictions: * The Api may be slow and take hundreds miliseconds. * Support only a very limited multicast ether address number. For instance of 25G Meallnox CX5, its PMD driver for rte_eth_dev_set_mc_addr_list supports only 128 multicast ether addresses for each port, and it takes about 250ms for rte_eth_dev_set_mc_addr_list to finish its work. Hence we make the following improvements: 1. Enable allmulitcast when rte_eth_dev_set_mc_addr_list failed, 2. Do not call rte_eth_dev_set_mc_addr_list when allmulitcast enabled. In next commmit, allmulticast can be configured at start of DPVS, which can solve the rte_eth_dev_set_mc_addr_list slow problem totally. Signed-off-by: ywc689 <[email protected]>
…vices Signed-off-by: ywc689 <[email protected]>
…packets out Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
when update vs/rs configuration parameters at keepalived reload, it's no need to delete and re-add vs/rs, which will result in traffic interruption of the vs/rs.
hardware multicast address bugfix
It works only for TCP by checking if syn-ack is replied from the backend, and if not, inhibit scheduling the backend for a period. The inhibiting period is gained/lost exponentially after successive checking failed/succeeded. Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
udp dest-check works only when the udp is bidirectional, that is, the udp conns that forwarding inbound only or outbound only are always detcted dead. Thus "dest-check" should be never configured for the single directional udp flow. Besides, the a smaller conn timeout may be specified for the bidirectional flow service to detect dest fault quickly. Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
fix #877 Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
Two versions -- v1-insecure and v2-insecure -- ared added for the proxy cascading case where the proxy protocol addresses should remain unchanged in the backend proxy server. Meanwhile, the v1 and v2 versions are always using the addresses from client's ip header of inbound packets. Signed-off-by: ywc689 <[email protected]>
1. Do not insert toa data in syn packets any more. 2. Compact tcp option space or remove some options when inserting toa data failed and then try again. 3. Emit a warning log for audit when toa insertion failed finally. Signed-off-by: ywc689 <[email protected]>
Signed-off-by: ywc689 <[email protected]>
ipvs: toa enhancements
Signed-off-by: ywc689 <[email protected]>
release v1.9.6
Features
Bugfixes
|
you-looks-not-tasty
approved these changes
Jan 2, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.