Unsolved
This post is more than 5 years old
1 Rookie
•
5 Posts
1
2328
November 13th, 2017 05:00
Strange output from Unity via uemcli
Hello,
If I give the following command in a dos box:
uemcli -d ***.***.***.*** -u Local/****** -p ******** -noHeader /sys/general show -filter "Health state"
I receive the following answer:
1: Health state = OK (5)
If I give the same command in a perl script then something went wrong, and I get the following answer:
■1 : H e a l t h s t a t e = O K ( 5 )
There are spaces between every character.
If I run the perl script and print the output in a file I see on every space the word NUL
ÿþ1NUL:NUL NUL NUL NUL NUL NULHNULeNULaNULlNULtNULhNUL NULsNULtNULaNULtNULeNUL NUL=NUL NULONULKNUL NUL(NUL5NUL) NUL NUL
I also use this script to check the health status on several VNX's without any problems.
Is there somebody who can help me with this?
Regards,
Efclem.
Rainer_EMC
4 Operator
•
8.6K Posts
0
November 13th, 2017 16:00
please open a service request and ask customer service if this warrants to file an engineering bug
efclem
1 Rookie
•
5 Posts
0
November 14th, 2017 03:00
Thanks, I opened a service request..
Rainer_EMC
4 Operator
•
8.6K Posts
0
November 14th, 2017 07:00
you're welcome
ChristianKrauss
2 Posts
0
September 18th, 2018 04:00
Hello, I have the same issue, did your service request resolved the problem? I think it is a encoding/UTF8? problem, so the output needs to be handled in Perl/Powershell.
ChristianKrauss
2 Posts
1
September 18th, 2018 06:00
Found the solution by myself, as it is documented in EMC Unisphere CLI Version 1.5.1 User Guide
Earlier versions of the CLI (prior to 1.5.1.1) used ASCII encoding or a superset of ASCII. In order to be able to display localized names, VNXe Unisphere CLI (Windows only) enforces the output to be in wide char format (UTF-16) with BOM. There have been reports of problems when using Perl on Windows with the VNXe Unisphere CLI. Since returned bytes are using an encoding that is not a superset of ASCII, this results in the characters being unreadable. The script writer may need to decode those bytes from UTF-16 into Unicode characters.
The following is an example of one way to decode the bytes within a Perl script. Note that the means of decoding in other scripting languages will vary:
if (@ans && $ans[0] =~ m/\x00/) {
my $s = decode('utf-16', join(encode('ucs-2', "\n"), @ans));
@ans = split(/\n/, $s);
}