Details
Description
This came up in the same context as MXS-4810, background is trying to run maxctrl in a shell script, fetching the output into a variable, basically:
result=$(maxctrl ... 2>&1)
|
if test $?
|
then
|
... error handling ...
|
fi
|
When hitting MXS-4810 a direct call of maxctrl leads to a non-zero error code being returned as expected, but when using maxctrl in a subshell to asign its output to a variable the error message changes, and the returned error code is zero, indicating success.
When running in a different error than MXS-4810, e.g. when only a single host is given and the 10 second timeout kicks in a non-zero error code is returned in both cases as expected.
Below is "how to reproduce" for both this and MXS-4810 as I originally mentioned in chat:
This is what I get when trying to contact a maxscale that does not actually exist:
$ time maxctrl list servers --hosts 172.29.7.245:8989
|
Error: timeout of 10000ms exceeded
|
|
|
real 0m10.275s
|
user 0m0.233s
|
sys 0m0.048s
|
but when trying with two host entries with no maxscale instance behind them it takes 2min10s instead of 10s, and the error message is different:
$ time maxctrl list servers --hosts 172.29.7.245:8989,172.29.7.246:8989
|
Error: connect ETIMEDOUT 172.29.7.246:8989
|
at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1159:16) {
|
errno: -110,
|
code: 'ETIMEDOUT',
|
syscall: 'connect',
|
address: '172.29.7.246',
|
port: 8989,
|
[... lots of additiona JSON lines ...]
|
response: undefined,
|
isAxiosError: true,
|
toJSON: [Function: toJSON]
|
}
|
|
|
real 2m9.818s
|
user 0m0.413s
|
sys 0m0.030s
|
in both cases the exit status in $? is non-zero at least, as expected
now lets try this in a subshell to fetch the returned text on stdout in a variable:
$ result=$(time maxctrl list servers --hosts 172.29.7.245:8989)
|
Error: timeout of 10000ms exceeded
|
|
|
real 0m10.226s
|
user 0m0.180s
|
sys 0m0.039s
|
|
|
$ echo $?
|
1
|
no surprise yet, but now with two hosts again:
$ result=$(maxctrl list servers --hosts 172.29.7.245:8989,172.29.7.246:8989)
|
(node:74204) UnhandledPromiseRejectionWarning: TypeError: undefined is not a function
|
at Object.strip_colors (/snapshot/maxctrl/lib/utils.js)
|
at print (/snapshot/maxctrl/maxctrl.js)
|
at err (/snapshot/maxctrl/maxctrl.js)
|
at processTicksAndRejections (internal/process/task_queues.js:95:5)
|
(Use `maxctrl --trace-warnings ...` to show where the warning was created)
|
(node:74204) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). To terminate the node process on unhandled promise rejection, use the CLI flag `--unhandled-rejections=strict` (see https://nodejs.org/api/cli.html#cli_unhandled_rejections_mode). (rejection id: 2)
|
(node:74204) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.
|
|
|
$ echo $?
|
0
|
so now getting a weird node exception instead, and worse: a SUCCESS exit code
i can understand the different timeout behavior if the application level timeout of 10s is not getting set up again a 2nd time before trying the next host entry
but I am at a total loss at why it would throw different errors when executed directly vs. within a subshell?
Attachments
Issue Links
- is caused by
-
MXS-4810 --timeout doesn't work with multiple values in --hosts
-
- Closed
-