Checkout Tools
  • last updated 6 hours ago
Constraints: committers
Constraints: files
Constraints: dates

Changeset 1854370 is being indexed.

backport proposal h2 slave keeapalives

backport proposal patch
    • ?
*) mod_http2/mod_proxy_http2: proxy_http2 checks correct master connection aborted status

to trigger immediate shutdown of backend connections. This is now always signalled

by mod_http2 when the the session is being released.

proxy_http2 now only sends a PING frame to the backend when there is not already one

in flight. [Stefan Eissing]

*) mod_proxy_http2: fixed an issue where a proxy_http2 handler entered an infinite

loop when encountering certain errors on the backend connection.

See <>. [Stefan Eissing]

Update proposal.

Add updated patch.

    • ?
fr doc rebuild.

XML fr doc files updates.

fr doc rebuild.

  1. … 3 more files in changeset.
XML fr doc files updates.

http: Fix possible empty response with mod_ratelimit for HEAD requests.

Don't eat the EOS in ap_http_header_filter() if it comes in single brigade

with a full response to a HEAD request, otherwise mod_ratelimit will never

flush its pending data.

Reviewed and corrected some spelling and typos

changed typo on "This module relies on OpenSSL to provide the cryptography engine." to change


mod_proxy_hcheck: mute "run from watchdog" message.

This is way too verbose (every 100ms) to be logged at any LogLevel.

Update a comment about the 'PATCH' HTTP command
Update mod_reqtimeout proposal to fit mod_http2.
proposal to include 1853967 in backport of mod_reqtimeout change, vote

mod_http2: fix for latest mod_reqtimeout change re handshake timeouts

Update proposal (and comment about it).
mod_proxy_http: revert spurious comment from r1853953.

No, if we read more bytes than the C-L there is really something wrong in

our (or some) HTTP input filter.

mod_proxy_http: common function for stream_reqbody_{cl,chunked}()

Since stream_reqbody_cl() and stream_reqbody_chunked}() now have the same

structure, join them into a single stream_reqbody() function which is passed

the rb_method to handle only CL vs chunked cases differently.

Follow up to r1853874: CHANGES entry.
mod_reqtimeout: follow up to r1853901: init stage name outside INIT_STAGE().

It helps both code readability where stages are initilized and potential

logging in the input filter (even if the stage is disabled).

mod_reqtimeout: follow up to r1853901: fix macro args.

mod_reqtimeout: follow up to r1853906: adjust hooks priorities comments.

mod_reqtimeout: Allow to configure (TLS-)handshake timeouts.

The timeouts apply between the process_connection and pre_read_request hooks.

They are disabled by default for compatibily reasons.

mod_reqtimeout: factorize structs and code.

With a bit of macro magic, this is to avoid more code duplication when adding

new stages (next commit will add TLS/handshake timeouts handling in addition to

existing header and body ones).

No functional change here.

mod_cache_socache: avoid pool to heap reallocation.

Below some threshold, the previous code tried free (sub-)pooled memory ASAP by

moving small buffers (< capacity / 2) to a heap bucket. But this is not really

an optimization because first it requires at some point to allocate more than

the configured capacity, and second since this happens during response handling

the pool is about to be destroyed soon anymay.

This commit simply keeps the data in the subpool and uses a pool bucket for the

output brigade to take care of the lifetime until it's consumed (or not).