RepoMirrors
/
haproxy
mirror of http://git.haproxy.org/git/haproxy.git/
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
haproxy/reg-tests/checks/http-check.vtc

158 lines
4.9 KiB
Plaintext
Raw Normal View History

REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
varnishtest "Health-checks: some http-check tests"
feature ignore_unknown_macro
#REQUIRE_VERSION=2.2
#REGTEST_TYPE=slow
# This script tests HTTP health-checks.
server s1 {
rxreq
expect req.method == OPTIONS
expect req.url == /
expect req.proto == HTTP/1.0
txresp
} -start
server s2 {
rxreq
expect req.method == GET
expect req.url == /status
expect req.proto == HTTP/1.1
txresp
} -start
server s3 {
rxreq
expect req.method == GET
expect req.url == /status
expect req.proto == HTTP/1.1
txresp
} -start
server s4 {
rxreq
expect req.method == GET
expect req.url == /req1
expect req.proto == HTTP/1.1
expect req.http.x-test == "server=srv"
expect req.http.x-haproxy-server-state ~ "UP.+name=be4/srv"
expect req.bodylen == 0
txresp
accept
rxreq
expect req.method == GET
expect req.url == /req2
expect req.proto == HTTP/1.1
expect req.http.x-test == "server="
expect req.http.x-haproxy-server-state ~ "UP.+name=be4/srv"
expect req.http.content-length == 17
expect req.bodylen == 17
expect req.body == "health-check body"
txresp
accept
rxreq
expect req.method == GET
expect req.url == /req3
expect req.proto == HTTP/1.0
expect req.http.x-test == <undef>
expect req.http.x-haproxy-server-state ~ "UP.+name=be4/srv"
expect req.bodylen == 0
txresp
MINOR: checks: Support log-format string to set the URI for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request URI. the keyword "uri-lf" should be used instead of "uri". If the string eval fails, we fall back on the default uri "/".
2020-05-06 13:06:34 +00:00
accept
rxreq
expect req.method == GET
expect req.url == /
expect req.proto == HTTP/1.0
expect req.http.x-test == <undef>
expect req.http.x-haproxy-server-state ~ "UP.+name=be4/srv"
MINOR: checks: Support log-format string to set the body for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request's body. the keyword "body-lf" should be used instead of "body". If the string eval fails, no body is added.
2020-05-06 13:38:58 +00:00
expect req.bodylen == 23
expect req.body == "health-check on be4-srv"
MINOR: checks: Support log-format string to set the URI for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request URI. the keyword "uri-lf" should be used instead of "uri". If the string eval fails, we fall back on the default uri "/".
2020-05-06 13:06:34 +00:00
txresp
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
} -start
syslog S1 -level notice {
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
BUG/MAJOR: tcpcheck: Allocate input and output buffers from the buffer pool Historically, the input and output buffers of a check are allocated by hand during the startup, with a specific size (not necessarily the same than other buffers). But since the recent refactoring of the checks to rely exclusively on the tcp-checks and to use the underlying mux layer, this part is totally buggy. Indeed, because these buffers are now passed to a mux, they maybe be swapped if a zero-copy is possible. In fact, for now it is only possible in h2_rcv_buf(). Thus the bug concretely only exists if a h2 health-check is performed. But, it is a latent bug for other muxes. Another problem is the size of these buffers. because it may differ for the other buffer size, it might be source of bugs. Finally, for configurations with hundreds of thousands of servers, having 2 buffers per check always allocated may be an issue. To fix the bug, we now allocate these buffers when required using the buffer pool. Thus not-running checks don't waste memory and muxes may swap them if possible. The only drawback is the check buffers have now always the same size than buffers used by the streams. This deprecates indirectly the "tune.chksize" global option. In addition, the http-check regtest have been update to perform some h2 health-checks. Many thanks to @VigneshSP94 for its help on this bug. This patch should solve the issue #936. It relies on the commit "MINOR: tcpcheck: Don't handle anymore in-progress send rules in tcpcheck_main". Both must be backport as far as 2.2. bla
2020-11-25 12:47:00 +00:00
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
recv
expect ~ "[^:\\[ ]\\[${h1_pid}\\]: Health check for server be[0-9]/srv succeeded.*code: 200"
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
} -start
haproxy h1 -conf {
defaults
mode http
timeout client 1s
timeout server 1s
timeout connect 100ms
option log-health-checks
backend be1
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk
server srv ${s1_addr}:${s1_port} check inter 100ms rise 1 fall 1
backend be2
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk GET /status HTTP/1.1
server srv ${s2_addr}:${s2_port} check inter 100ms rise 1 fall 1
backend be3
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk
MINOR: checks: Use ver keyword to specify the HTTP version for http checks 'ver' keyword is already used by sample fetches while 'vsn' is not used anywhere else. So better to use 'ver' too for http-check send rules.
2020-04-28 07:37:00 +00:00
http-check send meth GET uri /status ver HTTP/1.1
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
server srv ${s3_addr}:${s3_port} check inter 100ms rise 1 fall 1
backend be4
mode tcp
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk
http-check send-state
http-check connect addr ${s4_addr}:${s4_port}
http-check set-var(check.server) "str(srv)"
MINOR: checks: Support log-format string to set the URI for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request URI. the keyword "uri-lf" should be used instead of "uri". If the string eval fails, we fall back on the default uri "/".
2020-05-06 13:06:34 +00:00
http-check set-var(check.path) "str(/req1)"
http-check send meth GET uri-lf "%[var(check.path)]" ver HTTP/1.1 hdr x-test "server=%[var(check.server)]"
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
http-check expect status 200
http-check connect addr ${s4_addr} port ${s4_port}
http-check unset-var(check.server)
MINOR: checks: Support log-format string to set the URI for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request URI. the keyword "uri-lf" should be used instead of "uri". If the string eval fails, we fall back on the default uri "/".
2020-05-06 13:06:34 +00:00
http-check set-var(check.path) "str(/req2)"
http-check send meth GET uri-lf "%[var(check.path)]" ver HTTP/1.1 hdr x-test "server=%[var(check.server)]" body "health-check body"
http-check expect rstatus "^2[0-9]{2}"
http-check connect addr ${s4_addr} port ${s4_port}
http-check set-var(check.path) "str(/req3)"
http-check send meth GET uri-lf "%[var(check.path)]"
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
http-check expect rstatus "^2[0-9]{2}"
http-check connect addr ${s4_addr} port ${s4_port}
MINOR: checks: Support log-format string to set the URI for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request URI. the keyword "uri-lf" should be used instead of "uri". If the string eval fails, we fall back on the default uri "/".
2020-05-06 13:06:34 +00:00
http-check unset-var(check.path)
MINOR: checks: Support log-format string to set the body for HTTP send rules For http-check send rules, it is now possible to use a log-format string to set the request's body. the keyword "body-lf" should be used instead of "body". If the string eval fails, no body is added.
2020-05-06 13:38:58 +00:00
http-check send meth GET uri-lf "%[var(check.path)]" body-lf "health-check on %[be_name]-%[srv_name]"
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
## implicit expect rule
server srv ${s1_addr}:${s1_port} check inter 100ms rise 1 fall 1
BUG/MAJOR: tcpcheck: Allocate input and output buffers from the buffer pool Historically, the input and output buffers of a check are allocated by hand during the startup, with a specific size (not necessarily the same than other buffers). But since the recent refactoring of the checks to rely exclusively on the tcp-checks and to use the underlying mux layer, this part is totally buggy. Indeed, because these buffers are now passed to a mux, they maybe be swapped if a zero-copy is possible. In fact, for now it is only possible in h2_rcv_buf(). Thus the bug concretely only exists if a h2 health-check is performed. But, it is a latent bug for other muxes. Another problem is the size of these buffers. because it may differ for the other buffer size, it might be source of bugs. Finally, for configurations with hundreds of thousands of servers, having 2 buffers per check always allocated may be an issue. To fix the bug, we now allocate these buffers when required using the buffer pool. Thus not-running checks don't waste memory and muxes may swap them if possible. The only drawback is the check buffers have now always the same size than buffers used by the streams. This deprecates indirectly the "tune.chksize" global option. In addition, the http-check regtest have been update to perform some h2 health-checks. Many thanks to @VigneshSP94 for its help on this bug. This patch should solve the issue #936. It relies on the commit "MINOR: tcpcheck: Don't handle anymore in-progress send rules in tcpcheck_main". Both must be backport as far as 2.2. bla
2020-11-25 12:47:00 +00:00
backend be5
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk
server srv ${h1_li1_addr}:${h1_li1_port} proto h2 check inter 100ms rise 1 fall 1
backend be6
log ${S1_addr}:${S1_port} len 2048 local0
option httpchk GET /status HTTP/1.1
server srv ${h1_li1_addr}:${h1_li1_port} check check-proto h2 inter 100ms rise 1 fall 1
listen li1
mode http
bind "fd@${li1}" proto h2
http-request return status 200
REGTEST: Add scripts to test based tcp-check health-checks These scripts have been added to validate the health-checks based on tcp-check rules (http, redis, MySQL...).
2020-04-22 13:16:58 +00:00
} -start
syslog S1 -wait