Rocket® Open-Source Languages and Tools for z/OS®

I’ve just installed GNU bash, version 4.2.53(1)-release (i370-ibm-openedition) on Z/OS 2.3 and I’m somewhat confused by what config files bash uses at startup From …//share/doc/bash/4.3/README.ZOS the doc sartes Non-interactive mode The bash shell proceses configuration files in the following order: (1) /etc/bash.bashrc (2) ~/.bashrc Login-shell The bash login shell processes configuration files in the following order: (1) ~/.bash_profile (2) ~/.bash_login (3) ~/.profile Non-login shell The bash shell executes a file “~/.bashrc” The results for a non-login shell matches the doc However I do a: bash echo.sh (which I assume is a non-interactive shell) I see no config files executed If I try bash --login I onle see the follwoingt cofig file excured /etc/profile ~/…bash_profile

We have openssl installed but when I issue the following command results in no peer certificate and No client certificate CA names sent. Would this be because of the openssl not pointed to a certificate bundle? Command issued: openssl s_client -connect sfcertrequestTEST256.support.statefarm.org:443 ***** Results ****** CONNECTED(00000003) no peer certificate available No client certificate CA names sent SSL handshake has read 7 bytes and written 307 bytes New, (NONE), Cipher is (NONE) Secure Renegotiation IS NOT supported Compression: NONE Expansion: NONE No ALPN negotiated SSL-Session: Protocol : TLSv1.2 Cipher : 0000 Session-ID: Session-ID-ctx: Master-Key: Key-Arg : None PSK identity: None PSK identity hint: None SRP username: None Start Time: 1555530422 Timeout : 300 (sec) Verify return code: 0 (ok)

We are successfully using standalone Python and Python API in z/OS 2.2 except one thing. In standalone Python print() statements produce output that is correctly displayed on a screen (thanks to _BPXK_AUTOCVT=ON). However when we use API (with the same environment variables exported) the output from print() statements appears to be in ASCII. When this happens the stderr contains a line tput: unknown terminal “/pgm/u/myHomeDir” Our application is compiled as EBCDIC and switches environment to ASCII before using any API calls. Any idea what can be done to make print statements work? Thank you Best regards Jacob

Is there a date/ETA for when Postfix for z/OS is available for download? I have a pre-GA version and am unable to install it in our test environment, so I was wondering when the final version would be available.

I ran into problems downloading the .Tar file to the file system i created on z/OS. I created a file system with cyl(5000 150) When i TAR python-2017-04-12-py27.tar it run a long time and eventually runs out of space. I cant believe this file is actually this big. How big should my filesystem be? I followed your directions on installation of Rocket z/OS open source tools product. I appreciate your assistance.

Hi, I have some trouble moving curl for z/OS from version 7.16.1 to 7.52.1. We call curl from a program written in REXX (using bpxbatch routine) to call a websevice on a remote (non z/OS) server; this webservice returns data already encoded in EBCDIC. Using the old version the data is retrieved with the correct encoding, while the new version shows bad encoded data; I obtain the same behaviour even if I call curl from ISHELL or BPXBATCH. Please note that we need to get the data from the stdout without writing it to a unix file (If I use the -o option to write to a unix file the data is written with the correct encoding). This is what happens with curl for z/OS version 7.16.1: curl -v --header “Xml-Stylesheet:/xsl/tws-ws2zos.xsl” --user “***:***” “http://***” HPUC0330 About to connect() to *** port 80 (#0) HPUC0016 Trying ...… HPUC0017 connected HPUC0331 Connected to *** (...) port 80 HPUC0143 Server auth using Basic with user ‘***’ GET *** HTTP/1.1. Authorization: Basic *** User-Ag

I’ve installed Vim for z/OS. So far everything works as expected, except for that the Ctrl key mapping in my .vimrc is not working. for example, I have these key mappings in my .vimrc, and they are not working. map <C-d> :NERDTreeToggle<CR> map <C-j> <C-W>j map <C-k> <C-W>k map <C-h> <C-W>h map <C-l> <C-W>l

The following is for git repos on z/OS using git for z version 2.3.5. Our enterprise remote server is bitbucket v5.16.0 as of this writing. We are currently having encoding issues with bitbucket during PR merges; it is trying to encode from UTF-8 to ibm-1047 and failing, blocking any PR’s from being merged. An extract from our bitbucket logs: c.a.s.i.s.g.p.DefaultPullRequestAutoMergeStrategy CPDC/ci[6631]: Pull request 45@2 failed automatic merging: '/opt/git-2.19.0/bin/git merge -m Automatic merge --no-ff --no-log --allow-unrelated-histories cf734c8c560b13df7a76624b6296e4f75684f161' exited with code 0 saying: error: failed to encode 'NET#C/ciuba#c.cbl' from UTF-8 to ibm-1047 This issue was not present before and only began cropping up recently. Could this have something to do with the “working-tree-encoding” attribute now being used by distributed platforms like bitbucket server, and trying to convert to IBM-1047 on bitbucket server? We are currently running git 2.3.5 which does not h

It has been a long road today and I am going nowhere. I am reinstalling z/Open Source files after a install of z/OS 2.3. I strongly suspect that the tar file is being corrupted as it bounces around on its way to z/OS. Here are the steps I used. A download from the Rocket web site to my downloads file directory. A PC move from downloads to a desktop folder (for convenience) A Win/SCP BINARY (yes BINARY) transfer from my PC to a personal directory on z/os (/u/tech905.RocketOpen) The tar command is run (using the minimal instructions from the website AND from an example here in the forum) Both Get message tar: checksum error on tape (got 4355, expected 0) Of course, without gzip I cannot extract and test any of the other downloaded files. Is there a way to do a direct download from Rocket to z/OS? Where did this bouncing file get broken??

I have an API server running in my z/OS 2.3. If I use curl from OMVS to test the API I get the following error: /usr/lpp/ported/curl/bin>curl https://192.168.48.41:7050/zosConnect/apis -v -k --user ‘uuuuuuuu:pppppppp’ Trying 192.168.48.41… TCP_NODELAY set Connected to 192.168.48.41 (192.168.48.41) port 7050 (Ñ0) ALPN, offering h2 ALPN, offering http/1.1 Cipher selection: ALL:]EXPORT:]EXPORT40:]EXPORT56:]aNULL:]LOW:]RC4:@STRENGTH TLSv1.2 (OUT), TLS header, Certificate Status (22): TLSv1.2 (OUT), TLS handshake, Client hello (1): Unknown SSL protocol error in connection to 192.168.48.41:7050 Curl_http_done: called premature == 1 stopped the pause stream] Closing connection 0 curl: (35) Unknown SSL protocol error in connection to 192.168.48.41:7050 But if I use the curl in my Linux laptop, the response is correct: curl https://192.168.48.41:7050/zosConnect/apis -v -k --user ‘uuuuuuuu:pppppppp’ About to connect() to 192.168.48.41 port 7050 (#0) Trying 192.168.48.41… Connected to 192.

Does anyone know how to open a MVS dataset in python 3.6.1? And how to refer to a DD name in JCL? I have tried using open("//‘xxx.xxx.xxx’) and it returned ‘FileNotFoundError’ but the file is definitely exist! I can access the file in USS without any problem.

Hi, We have installed GIT and the corequisite products for z/OS we are running version 2.12.0 bash-4.3$ git --version git version 2.12.0 It appears that I can create a local repo and it works ok, I can add code to it and it appears to track changes. But whenever I try to pull code from the Remote Repo I get the following error. bash-4.3$ git clone https://nbkbjwz@horizon.bankofamerica.com/scm/scm/bcc/bcc.git Cloning into ‘bcc’… fatal: bad config line 1 in file /u/user/nbkbjwz/test9/bcc/.git/config I believe the problem is related to the Bitbucket repo using UTF-8 we have all parameters set correctly and the Bash environment is established when we login to SSH. Using booth the profile and bash RC scripts. bash-4.3$ set BASH=/bac/usr/BOFA/git/bin/bash BASHOPTS=cmdhist:complete_fullquote:expand_aliases:extquote:force_fignore:hostcomplete:interactive_comments:progcomp:promptvars:sourcepath BASH_ALIASES=() BASH_ARGC=() BASH_ARGV=() BASH_CMDS=() BASH_LINENO=() BASH_SOURCE=() BASH_VERSINFO=

Hello, I’ve downloaded the Python files. I’m unsure what my next steps are. Do I upload these files into my USS file system someplace? I’m guessing that I have to create an aggregate to hold the files, not installed into root. How do I make the install, changes.zos and readme.zos files readable and formatted? I see in other questions that I might want to download ‘other’ software. Is the complete list of the apps I need documented someplace?

Hello, Would someone know how to get the following data? The environment is zOS mainframe. My programming language is REXX Language for zOS mainframe. Through Rexx language, it’s possible to get data coming from System Resource Manager Resource Control Table. For example, From RCTCLAS field we can get information about Long-term average CPU service used by this logical partition, in millions of service units per hour. From RCTIMGWU field we can get information about Workload Units available to MVS image when not running as VM guest From RCTCECWU field we can get information about workload units capacity by CEC. These data are related to one specific LPAR environment. It’s possible to get more detail about these fields reading IBM zOS MVS Data Areas manuals. Now, I am trying to get information of two data about LPAR GROUP MSU/H in use at any time. Average Unused MSUs/hour. I have tried to find out how to get these data reading the IBM MVS Data Areas manual but I haven’t been able to g

I am attempting to PUT a z/OS load module on a Linux system into a repository to be pulled later and run on z/OS. I am using curl -X PUT -k -u userid:passowrd -v -T --data-binary @/folder/Jzosfile.so “https://dnsname/path/filename” I get an error “curl: Can’t open ‘–data-binary’!”. Is this not a valid option on PUT? I see people asking questions about cURL POST with --data-binary but not PUT. Any suggestions on how to put a file in binary from cURL? I do it all the time FTP but what I am putting this do does not support FTP. Thank you

Hi, Can we use centralized repositories like on premesis Git server or bitbucket instead of using GitHub for git on Z/OS platform? Danish

Can somebody help me with the below: Is it possible to integrate IBM DBB with Endevor? If I host Git, can DBB use Endevor libraries in builds for doing Devops?

Hi everybody, I am trying to use Python statistical functions and getting the following error: raise NotImplementedError(‘cannot determine number of cpus’) that’s coming from the file /usr/lpp/python36/lib/python3.6/multiprocessing/context.py specifically from the lines 40-46 num = os.cpu_count() if num is None raise NotImplementedError(‘cannot determine number of cpus’) else: return num Is there a way to make statistical packages work? Or make os.cpu_count() return 1? Thanks in advance UPDATE: I was able to override os.cpu_count() with the function that always returns ‘1’. So multiprocessing can initialize now and stats packages seem to work now.

Hi, Is it possible to execute a MVS console commmand (ie, D IPLINFO) from a Bash script and get the output? I know I can use Java or even REXX, but plain old Unix is much easier for me Thanks!

Just a simple question. When DBtools 4.4.0 is going to be released? or actually, what other tools can we use to work with Unidata in an environment where the Unidata server is remote and shared with a lot of developers? And now that we have Python, what can we use to develop also in this new language inside Unidata? Thanks in advance

Not sure if this is a problem with our configuration or a problem with Git. But any git stash command is failing as seen here: gitstash.png931×338 5.43 KB This is with git version: 2.14.4_zos_b08 Anyone else having the same problem ? Regards, Gary

Our mainframe does not have direct access to the internet. Does anyone have a recommendation on the best way to close a github repository to z/OS with this being the case?

I’m attempting to open an MVS file with Python. I can do a: cat “//‘XXX.YYY’” and display the file just fine. However, I’m having trouble opening the file with Python. Pretty sure it has something to with needing to ESCAPE some of the characters in the file name Any ideas? Thanks,

I recently got Ansible working on zOS thanks to Rocket’s port of python and am looking to upload some ‘stuff’ to ansible galaxy (https://galaxy.ansible.com/) for people to use as an example. So far I’m thinking about uploading ‘cowsay’ but it would be nice if I uploaded something a little more useful. Would Rocket be opposed to having some of the OSS they have provided available on ansible galaxy? See here for discussion - https://github.com/ansible/ansible/issues/18555 I was thinking about starting with Rocket’s port of git and all of it’s dependencies. Edit: I also see there is an ansible-conda module (https://github.com/UDST/ansible-conda)

https://forum.rocketsoftware.com/t/colors-are-not-interpreted-correctly-when-using-git-diff/612 Similar to the above post, as we do not have less installed at our shop, I was looking for alternative pagers to set for git core.pager. I came across “more” and “pg”. When I set more as the pager, I get all sorts of funky output which seems to be “more” not interpreting the colors correctly(see below). image.png1164×225 10.4 KB When using ‘pg’ as the pager, the log looks like this: image.png1227×180 8.14 KB Is there any way to configure “more” so that it interprets colors correctly? I really like the scrolling functionality, pg is not intuitive to use.