Rocket Mainframe DevOps

I have been working test automation that uses screen to keep processes running while the driving process lets go of the terminal. I no longer what to continue to use nohup on z/OS for similar function. As z/OS is the last platform in my mixed environment to provide screen, I figured it would be worth asking Rocket if this were something worth tackling. https://www.gnu.org/software/screen/ If I had to give it a value, Lets say it would enable a use to reconnected to an existing session after being disconnected. Sort of like ISPF does today with reconnect.

Status: DeclinedTo amend the discrepancy report in Changeman SSM, to add the package id information for BASE and PRODUCTION details.For the Change man SSM Discrepancy report, we would like to see under below format: BASE0 component name, its timestamp and its corresponding package id PROD component name, its timestamp and its corresponding package id Discrepancy details – similar to comparex report The item #1 and #2 preferably appears on the same line. Justification: Since all codes are baseline/implemented via Changeman package only it make sense to include the package ID information for all un-synch code components. CMAN SSM report will contribute more value add if can show complete discrepancy information to our Developer and Prod support team. It will provide aide to our Dev and Prod stakeholders to identify from which library the correct component version resides thus generate savings in terms of time and effort as staff resources are utilize in an efficient manner

We are currently experimenting with using Jenkins remote slave on USS to issue Git commands to Git installation. We have two test systems. On one of the test systems everything works as expected. On the second test system the Jenkins build fails during the Git fetch because it cannot find any remote branches. Both test systems are being invoked by the same Jenkins server and are connecting to the same remove Git server. We have not been able to identify what if any differences between the systems may be causing the inconsistent behavior. We are using a standard Jenkins Git Client plugin to access the remote repository. It issues four commands at the beginning of each build. Jenkins Build Log from the failing test system: Building remotely on tvt6031Dan in workspace /u/dbruce/test/jenkins/workspace/DansGitPull > /u/dbruce/test/dbb/sample/git-jenkins.sh rev-parse --is-inside-work-tree # timeout=10 Fetching changes from the remote Git repository > /u/dbruce/test/dbb/sample/

SQLite would be a great addition to your Open Source Tools. The current version (3.25.2) should compile without any changes and it would be a great back-end DB when a full blown Database is overkill.

Git version 2.14.4 for z/OS is now available for download from Rocket Software, at: https://my.rocketsoftware.com/ This version supersedes the prior version, Git 2.3.5. In addition to the platform-generic changes from the previous version, the following z/OS-specific changes have been made: The https protocol is now supported for access to remote repositories, in addition to the ssh protocol. The working-tree-encoding attribute has been replaced by a new attribute, zos-working-tree-encoding. NOTE It is vital that all users of the working-tree-encoding attribute migrate to the new zos-working-tree-encoding attribute as soon as possible. See the release notes notes (supplied with the download) regarding an incompatibility with git releases after 2.18 on other platforms. Git for zOS 2.14.4 - Release Notes.pdf (63.3 KB)

Hi, I am getting the following error while trying to upgrade PIP on Unix on z/os. Our admin is saying that I have the authority to bin folder. Thanks in advance for any help. $ python -m pip install --upgrade pip Collecting pip Using cached https://files.pythonhosted.org/packages/5f/25/e52d3f31441505a5f3af41213346e5b6c221c9e086a166f3703d2ddaf940/pip-18.0-py2.py3-none-any.whl pip utils init unpack_filename: unpacked: /SYSTEM/tmp/pip-mdslhlp5-unpack/pip-18.0-py2.py3-none-any.whl Installing collected packages: pip Found existing installation: pip 9.0.1 Uninstalling pip-9.0.1: Exception: Traceback (most recent call last): File “/usr/lpp/python/python-2017-04-12-py36/python36/lib/python3.6/shutil.py”, line 551, in move os.rename(src, real_dst) OSError: [Errno 144] EDC5144I Improper link.: ‘/usr/lpp/python/python-2017-04-12-py36/python36/lib/python3.6/site-packages/pip-9.0.1.dist-info/DESCRIPTION.rst’ -> ‘/tmp/pip-fxzllj_9-uninstall/usr/lpp/python/python-2017-04-12-py36/python36/lib/pytho

Hi All I obtain an unusable output of STDOUT from my batch zos job with the UNZIP tool Thanks to all to help me Best Regards Christian LAYEC French User of Shadow z/Services //BPXBATCH EXEC PGM=IKJEFT01,REGION=0M,TIME=NOLIMIT //SYSTSPRT DD SYSOUT=* //STDOUT DD SYSOUT=* //STDERR DD SYSOUT=* //SYSTSIN DD * BPXBATCH SH + export PATH=$PATH:/usr/lpp/produits/unzip/bin ; + export MANPATH=$MANPATH:/usr/lpp/produits/unzip/man ; + export _BPXK_AUTOCVT=ON ; + export _CEE_RUNOPTS=‘FILETAG(AUTOCVT’ ; + export _TAG_REDIR_ERR=txt ; + export _TAG_REDIR_IN=txt ; + export _TAG_REDIR_OUT=txt ; + cd /usr/lpp/produits/unzip/temp ; + unzip -l SUPKR00P.ZIP ; /* READY BPXBATCH SH export PATH=$PATH:/usr/lpp/produits/unzip/bin ; export MANPATH=$MANPATH:/usr/lpp/produits/unzip/man ; export _BPXK_AUT OCVT=ON ; export _CEE_RUNOPTS=‘FILETAG(AUTOCVT’ ; export _TAG_REDIR_ERR=txt ; export _TAG_REDIR_IN=txt ; export _TAG_REDIR_OUT=txt ; cd /usr/lpp/produits/unzip/temp ; unzip -l SUPKR00P.ZIP ; READY END ÊÄÇÑÎ

It can sometimes be useful to run git commands in batch from JCL. This turns out to be a bit tricky; here’s an approach that works for me. I would be interested in hearing about other approaches. The key documentation for this is UNIX System Services User’s Guide: The BPXBATCH utility. There are several issues that make this challenging: There are some unpleasant restrictions on the handling of the standard I/O streams, stdin/stdout/stderr: All three standard I/O streams are, by default, directed to /dev/null. stdin cannot be directed to an MVS dataset. It may be directed to a file in the USS file system. stdout and stderr may be directed to MVS datasets, but you have to be careful about the DCB characteristics. The command to be executed is passed to BPXBATCH via the PARM option on the EXEC statement; this is limited to a maximum length of 100 characters. My JCL for running BPXBATCH looks like this: //STEP1 EXEC PGM=BPXBATCH //STDOUT DD SYSOUT=* //STDPARM DD * SH command_t

Hello, When I was installing Python 2.7.13 and Python 3.6.1 for z/OS, I hit the same issue. Below I take the error for Python 3.6.1as example. Ran python36/bin/install_all_packages, the error is as follpwing. BASH_PREFIX is /bin PERL_PREFIX is /usr/bin PKGS_BASE is /u/mvsbuild/py361z/pkgs installing from /u/mvsbuild/py361z/pkgs into /u/mvsbuild/py361z/python36 /u/mvsbuild/py361z/pkgs/.install installing 0 files from /u/mvsbuild/py361z/pkgs/.install to /u/mvsbuild/py361z/python36 pax: ▒▒>/▒▒?▒/Ŏ▒▒>▒>▒▒/%^/%^▒/▒,/▒▒ˎ▒▒>▒>▒▒/%^▒/▒,/▒▒▒▒>▒>▒▒/%^▒▒▒▒▒^▒▒▒▒^▒/▒▒/▒▒ˎ▒▒>▒>▒>▒▒/%^▒/▒,/▒▒▒>▒▒▒▒/%?▒/ǎ▒>▒▒▒▒/▒▒?▒/▒^▒>▒?Ď▒>▒▒▒▒▒>▒?▒▒^▒▒/>▒/▒▒^▒▒▒▒Ď▒>▒▒▒▒▒>▒?▒▒^▒▒/>▒/▒▒^▒▒▒▒ǎ▒>▒▒▒▒▒▒▒^▒>▒^▒▒>▒▒▒?>▒ǎ▒>▒▒▒▒▒▒▒▒▒▒^▒▒▒▒▒▒▒▒▒/▒▒▒▒ǎ▒>▒▒▒▒▒▒▒▒^▒▒▒▒▒/▒^▒/▒▒▒▒▒Ď▒>▒▒▒▒▒▒▒▒^▒▒▒▒▒/▒^▒/▒▒▒▒▒ǎ▒>▒▒▒▒▒▒▒▒^▒▒/>▒▒▒Ď▒>▒▒▒▒▒▒▒▒^▒▒_/▒▒?▒▒Ď▒>▒▒▒▒▒▒▒▒^▒▒_/▒▒?▒▒ǎ▒>▒▒▒▒▒▒▒▒^:?▒Ď▒>▒▒▒▒▒▒▒▒^:?▒ǎ▒>▒▒▒▒▒▒_/▒▒?▒▒ǎ: EDC5129I No s

I’ve installed and used python in my own home area for some time now. To make python available for my colleagues I’ve moved it to a common folder called /sh/utils. When python permission is set to 755 then my colleagues get an error, but if I change permission to 775 (they are member of same group as I am) then it works fine. I would like to spread it out to other colleagues that are not in the same group and I don’t want to set permission to 777. Can users somehow set up an environment variable to point to their own home directory somewhere?

ZMF RACF PassTicket Generation ErrorsSteve Nevin walks us through troubleshooting RACF PassTicket generation errors#Videos#Announcement#ChangeManZMF#HowTo-BestPractice

Hello, I was curious if any of the ported tools embed a copy of an EBCDIC-safe bcrypt library? bcrypt itself is quite small, but has some ascii-isms. Thanks!

In z/OS 2.2, I am using: git version 2.3.5_os390_b013 /bin/man as shipped by IBM When I type “man git”, the second paragraph of the DESCRIPTION section gets truncated, such as the second line of the example: “See gittutorial(7) to get started, then see Everyday Git[1] for a useful minimum set of commands. The Git Userb After you mastered the basic concepts, you can come back to this page to learn what commands Git” There are other sections like this as well. If I do a hex dump on the man page, I see that there are instances where a x’0019’ combination of bytes appears in various places where the truncation occurs. I issue this command; these are IBM’s man & od commands: $ man git | od -t cxC Which shows the binary characters after the string “The Git Userb”: 0000001420 T h e G i t U s e r b \\0 031 40 E3 88 85 40 C7 89 A3 40 E4 A2 85 99 82 00 19 Is the git man page shipped this way, or did I miss something in my install?

Anyone know what I am doing incorrect? Everything works fine, if I perform the commands manually on USS and it works fine off z/OS. It doesn’t appear to be a security issue. It appears when the git clone command runs, it is not in the correct directory for the .git repo. Thanks, Jerry Building remotely on zOStest (zOS) in workspace /usr/lpp/tools/jenkins/workspace/GradleTestJava Cloning the remote Git repository Cloning repository ssh://git@xxxx:port/ged/gradletestjava.git /usr/lpp/tools/git/bin/git init /usr/lpp/tools/jenkins/workspace/GradleTestJava # timeout=10 Fetching upstream changes from ssh://git@xxxx:port/ged/gradletestjava.git /usr/lpp/tools/git/bin/git --version # timeout=10 /usr/lpp/tools/git/bin/git --version # timeout=10 using GIT_ASKPASS to set credentials /usr/lpp/tools/git/bin/git fetch --tags ssh://git@xxxx:port/ged/gradletestjava.git +refs/heads/:refs/remotes/origin/ ERROR: Error cloning remote repo ‘origin’ hudson.plugins.git.GitException: Command “/usr/lpp/tool

Hello, I am trying to install the 5.24 version for our developers and since this is a first for me, as Perl was always apart on IBM Ported tools… I have followed the doc and added PATH, LIBPATH and PERL5LIB. But everytime they goto execute PERL. As soon as they hit the first perl statement, they get an error like strict for example. for example: make>perl ./makeall.pl -Nut Tue Aug 14 08:40:45 2018 Building makefiles… use: /public/dev/b24/users/jkxxxx/acwa/jk21/make/bldmakefile.pl 107: FSUM7351 n ot found use: /public/dev/b24/users/jkxxxx/acwa/jk21/make/bldmakefile.pl 109: FSUM7351 n ot found use: /public/dev/b24/users/jkxxxx/acwa/jk21/make/bldmakefile.pl 111: FSUM7351 n ot found /public/dev/b24/users/jkxxxx/acwa/jk21/make/bldmakefile.pl 112: FSUM7332 syntax error: got (, expecting Newline makeall.pl: Error building makefiles I can only assume I did something wrong along the way? Can anyone give me hint were I went wrong?

Hi, I am using CURL on z/OS including in my CPI the Userid and PSW. I want to use digital certificates, How can I do it? I am using CURL to access to z/OSMF as my REST Server. Thanks,

Hello I am a new user of R and I am stuck. I want to execute a Rscript that will allow me to compute some indices for many stations once. Here is what the manual says to do: "5.2 Calculating the indices on multiple station files The script that provides this functionality is climpact2.batch.stations.r. This script requires command line arguments to be passed to it at run time. Execution of this script takes the following form, from the Linux command line: Rscript climpact2.batch.stations.r /full/path/to/station/files/ /full/path/to/metadata.txt base_period_begin base_period_end cores_to_use The 5 command line arguments above are defined in the following table. Table 2. Command line arguments to pass to climpact2.batch.stations.r /full/path/to/station/files/ Directory where individual station files are kept. An example can be found in sample_data/XXXX /full/path/to/metadata.txt Text file that contains information about each station file to process. base_period_begin Beginning year for t

Just upgraded a 4 system sypslex from curl 7.16.1 to curl 7.52.1. My systems share binaries and BPXPRM members so all is the same - we have AUTOCVT=NO On 1 system in the sysplex when I either use putty or omvs the curl responses are gibberish and seem to be in ASCII. However … on the other 3 systems the response is readable and in EBCDIC. Keep in mind this problem is also seen when just doing curl --help. If I issue export _BPXK_AUTOCVT=ON … then the issue is resolved and text response is readable. I know the documentation says all sessions should have the env variables set … but does anyone know why 1 of my 4 systems in the sysplex is defaulting differently? I opened an SR with IBM but they said to reach out to ROCKET.

Hello: Just downloaded cURL and when I try my first test I get a S0C1 abend. Is there a machine level dependency? We are running a Z10. Thanks

Hi, I’m trying to use “git bisect” on z/OS, but it crashes with a segmentation fault: $ git version git version 2.3.5_os390_b013 $ git bisect start HEAD 4471995 -- CEE5203S The signal SIGSEGV was received. error: git-bisect died of signal 11 Is anybody using “git bisect” successfully? Is this a known problem?

Hello! I couldn’t find any similar questions on this in the forum so I’ve posted my own. I recently installed perl 5.24 on zOS. When we use the perldoc command, for example ‘perldoc perlbot’, the command returns garbage characters along with valid text, like so: ^[1mNAME^[m perlbot - Links to information on object-oriented programming in Perl ^[1mDESCRIPTION^[m For information on OO programming with Perl, please see perlootut and perlobj. (The above documents supersede the collection of tricks that was formerly here in perlbot.) /tmp/D2m_sMWEA1 I found what I think are the source files for these pages in /share/man/cat1. They look OK there. The output looks mostly ok but not perfect if I’m in the OMVS shell (called from TSO). It looks worse in putty, which is what I was using in the example above. I’m wondering what I did wrong on the install/configuration. I do have a file not found that didn’t pass the check after the install : /lib/perl5/5.24.0/os390/perllocal.pod . It gets a ‘f

Hello, I am hoping to find a port of Xvfb for z/OS 2.3 that I can download and install. I didn’t find it on the current Rocket OPENSOURCE downloads list. Regards Bruce Hewson

When I try to look at a manpage from openssl, e.g. man sha256 I get: FSUMF061 Cannot format /usr/local/ssl/man/man1/sha256.1 Using the file command: file /usr/local/ssl/man/man1/sha256.1 I get: /usr/local/ssl/man/man1/sha256.1: nroff, eqn, or tbl text input text When scanning the rocket ported tools for files which are nroff, eqn, or tbl text input text I see that most if not all man pages from php, openssl, libssh2 are not usable. A possible fix from my side is to dowload the man page to my laptop (using ascii conversion) and run something like: mv openssl.1 openssl.1.bad groff -mandoc -Tascii openssl.1.bad > openssl.1 and then upload to the z/OS host.

apache-ant’s VERSION.ZOS looks like follows: Tool: apache Version: ant-1.9.2 Build Number: 0002 Shouldn’t it be? Tool: apache-ant Version: 1.9.2 Build Number: 0002

I’m installing Perl under USS on z/OS 2.2 to support Python for a RedHat Ansible PoC. Everything was going swimmingly until I tried to run ./change_pwd_perl.sh, during which I received a fail for the file …/lib/perl5/5.24.0/os390/perllocal.pod (output from the script is attached). I looked in the directory and the perllocal.pod file wasn’t there. Is this normal, or am I missing something?change_pwd_perl_sh_output.docx (12.2 KB)