Update awscli tests to log error messages #153
Conversation
nitisht
force-pushed
the
awscli-error-log
branch
from
62a91bd
to
d76df31
Compare
run/core/awscli/test.sh
Outdated
| if [ "$HASH_1_MB" == "$hash2" ]; then | ||
| function="delete_bucket" | ||
| out=$(delete_bucket "$bucket_name") | ||
| rv=$? | ||
| # remove download file | ||
| rm -rf /tmp/datafile-1-MB |
There was a problem hiding this comment.
Just rm -f is better -r is recursive and we can avoid recursive if not needed..
run/core/awscli/test.sh
Outdated
| else | ||
| rv=1 | ||
| out="Downloaded file verfication failed" |
There was a problem hiding this comment.
Verification failed for downloaded object
run/core/awscli/test.sh
Outdated
| @@ -354,6 +366,7 @@ function test_copy_object() { | |||
| rv=$? | |||
| else | |||
| rv=1 | |||
| out="Copied file verfication failed" | |||
There was a problem hiding this comment.
Verification failed for copied object
run/core/awscli/test.sh
Outdated
| @@ -158,6 +161,7 @@ function test_create_object() { | |||
| rm -rf /tmp/datafile-1-MB | |||
| else | |||
| rv=1 | |||
| out="Created object hash verification failed" | |||
There was a problem hiding this comment.
Checksum verification failed for uploaded object
run/core/awscli/test.sh
Outdated
| @@ -158,6 +161,7 @@ function test_create_object() { | |||
| rm -rf /tmp/datafile-1-MB | |||
There was a problem hiding this comment.
s/test_create_object/test_upload_object/
nitisht
force-pushed
the
awscli-error-log
branch
2 times, most recently
from
faeff5b
to
9e4454a
Compare
|
updated @harshavardhana |
There was a problem hiding this comment.
- Can we remove the /tmp/multipart file once it is used?
- The buckets are not deleted after a failure in the code, since there is no exception, we should try to clean it up.
- Also, the error field should be used for stack trace alone, so, do you think we should change the error field here to message
- If AWS returns an error output, they need to be converted to a single line?
Here is an example
{"name": "awscli", "duration": "623", "function": "aws --endpoint-url http://127.0.0.1:9000 s3 sync s3://awscli-mint-test-bucket-6887/\n", "status": "FAIL", "error": "usage: aws [options] <command> <subcommand> [<subcommand> ...] [parameters]
To see help text, you can run:
aws help
aws <command> help
aws <command> <subcommand> help
aws: error: the following arguments are required: paths"}
- Should we log the entire awscli command here in the case of function ? Will it be useful to search if the function contains actual endpoint information, bucket names etc? Or can we simplify the function value to just what we are trying to do here, that way it is consistent between endpoint A and endpoint B.
nitisht
force-pushed
the
awscli-error-log
branch
from
9e4454a
to
c017767
Compare
Done
Added bucket removal logic to error logging cases.
Removed few of the cases. But in cases where
done
I understand this will be taken care based on the discussion we had during today's call. |
|
@balamurugana found one more issue with this log. The duration is a number and it should not have double quotes around the log. |
nitisht
force-pushed
the
awscli-error-log
branch
from
c017767
to
25b6735
Compare
Fixed |
Fixes #150