Merge ~rafaeldtinoco/ubuntu/+source/crmsh:lp1895728-groovy into ubuntu/+source/crmsh:ubuntu/devel

[Rafael David Tinoco]

BUG: https://bugs.launchpad.net/ubuntu/+source/crmsh/+bug/1895728
PPA: https://launchpad.net/~rafaeldtinoco/+archive/ubuntu/builds

[Rafael David Tinoco (rafaeldtinoco)]

I'm bringing released fixes to crmsh because it is our tool of choice for cluster management (until PCS becomes the chosen one). I have tested this change with autopkgtest locally and had faced issues just because autopkgtest brakes the testbed (you have to completely uninstall corosync/pacemaker/crmsh and install them again in order to have a good test run).

[Rafael David Tinoco (rafaeldtinoco)]

Good for review whenever someone can. Thank you!

[Bryce Harrington (bryce)]

* Changelog:
  - [√] old content and logical tag match as expected
  - [√] changelog entry correct version and targeted codename
  - [√] changelog entries correct
  - [√] update-maintainer has been run

* Actual changes:
  - [ ] no upstream changes to consider
  - [√] no further upstream version to consider
  - [√] debian changes look safe

* Old Delta:
  - [-] dropped changes are ok to be dropped
  - [-] nothing else to drop
  - [-] changes forwarded upstream/debian (if appropriate)

* New Delta:
  - [-] no new patches added
  - [√] patches match what was proposed upstream
  - [√] patches correctly included in debian/patches/series
  - [√] patches have correct DEP3 metadata

* Build/Test:
  - [ ] build is ok
  - [ ] verified PPA package installs/uninstalls
  - [ ] autopkgtest against the PPA package passes
  - [ ] sanity checks test fine

I didn't take time to run the tests or try the installation, but can do so tomorrow if desired. I assume if there were test issues, they'll show up in the autopkgtest runs.

* Code review on patches:
  - ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch
    + Code changes looks correct & safe
    + Fix Origin: url is incomplete (needs SHA appended)
    + "bsc#1166967" sounds like a bug tracker ticket but I could not find a public bug system with that identifier. If there is one, it would be useful to link to it via a "Bug: ..." DEP3 tag. From the upstream commit entry there's a link to the pull request which has some discussion, so lacking access to the 'bsc', you could add "Bug: https://github.com/ClusterLabs/crmsh/pull/545".

  - ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch
    + Fix Origin: url is incomplete (needs SHA appended)
    + Add Bug: https://github.com/ClusterLabs/crmsh/pull/486 (or bsc#1158060)
    + It looks like the important piece of this patch is dropping the "echo t > /proc/sysrq-trigger" line. However it is also adding dump_D_process() to add some process state info; dump_D_process() is running a few bash commands which feels a bit janky. Since this is not an SRU I think it's good enough, but if it was an SRU I'd suggest dropping dump_D_process() and carrying only the actual fix.

  - ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch
    + LGTM. Personally I'd use $() rather than `` but this is technically correct and using mktemp is a good improvement.
    + Fix Origin: url is incomplete (needs SHA appended)
    + Add Bug: https://github.com/ClusterLabs/crmsh/pull/502 (or bsc#1154163)

  - ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch
    + Fix Origin: url is incomplete (needs SHA appended)
    + Add Bug: https://github.com/ClusterLabs/crmsh/pull/542 (or bsc#1166684)
    + Code changes look good per the pull request, although I wonder if csync2_update() should throw an exception if the second `csync2 -rxv` fails, but the warning is at least an improvement over the prior code.
    + Just a nitpick but I notice this patch uses "synced" in one message and "sync'd" in another.

  - ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-s...

review: Needs Fixing

[Rafael David Tinoco (rafaeldtinoco)]

> I didn't take time to run the tests or try the installation, but can do so
> tomorrow if desired. I assume if there were test issues, they'll show up in
> the autopkgtest runs.

I can test that Bryce, thanks a lot for reviewing this for me!

>
> * Code review on patches:
> - ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-
> sbd-re.patch
> + Code changes looks correct & safe
> + Fix Origin: url is incomplete (needs SHA appended)
> + "bsc#1166967" sounds like a bug tracker ticket but I could not find a
> public bug system with that identifier. If there is one, it would be useful
> to link to it via a "Bug: ..." DEP3 tag. From the upstream commit entry
> there's a link to the pull request which has some discussion, so lacking
> access to the 'bsc', you could add "Bug:
> https://github.com/ClusterLabs/crmsh/pull/545".

YES! I forgot about bsc# numbers and the original bug reference! These are: https://bugzilla.suse.com/show_bug.cgi?id=XXXXX as crmsh is maintained mostly by SUSE. Unfortunately they don't keep opened bug reporting systems like we do, so some of those references are not worth adding as there is noting to read. Ill try to replace dos with github issues if appropriate.

I have also forgotten to update the Origin with the commit SHAs. Doing that..

>
> - ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-
> dmes.patch
> + Fix Origin: url is incomplete (needs SHA appended)
> + Add Bug: https://github.com/ClusterLabs/crmsh/pull/486 (or bsc#1158060)
> + It looks like the important piece of this patch is dropping the "echo t
> > /proc/sysrq-trigger" line. However it is also adding dump_D_process() to
> add some process state info; dump_D_process() is running a few bash commands
> which feels a bit janky. Since this is not an SRU I think it's good enough,
> but if it was an SRU I'd suggest dropping dump_D_process() and carrying only
> the actual fix.

Yep, I don't find nice as well .. things like get_stdout("cat /proc/XXX").. I see your point and agree. For stabilization is good enough, as you said.

> - ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-
> bsc-1.patch
> + LGTM. Personally I'd use $() rather than `` but this is technically
> correct and using mktemp is a good improvement.
> + Fix Origin: url is incomplete (needs SHA appended)
> + Add Bug: https://github.com/ClusterLabs/crmsh/pull/502 (or bsc#1154163)

Alright.

> - ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-
> bsc-1166.patch
> + Fix Origin: url is incomplete (needs SHA appended)
> + Add Bug: https://github.com/ClusterLabs/crmsh/pull/542 (or bsc#1166684)
> + Code changes look good per the pull request, although I wonder if
> csync2_update() should throw an exception if the second `csync2 -rxv` fails,
> but the warning is at least an improvement over the prior code.
> + Just a nitpick but I notice this patch uses "synced" in one message and
> "sync'd" in another.

Yep :\. I tend not to change upstream unless we face the bug ourselves and suggest the changes. To be honest, bootstrap is something from crmsh we ...

[Rafael David Tinoco (rafaeldtinoco)]

Okay, finished replying to your review, will force push and ask for a second quick look and then upload if you are +1 tomorrow. Thanks again Bryce!

[Bryce Harrington (bryce)]

With Bug/Origin DEP3 tags fixed, giving it my +1.

review: Approve

[Rafael David Tinoco (rafaeldtinoco)]

[rafaeldtinoco@groovy crmsh]$ git describe
upload/4.2.0-3ubuntu1

[rafaeldtinoco@groovy crmsh]$ git push pkg upload/4.2.0-3ubuntu1
Enumerating objects: 27, done.
Counting objects: 100% (27/27), done.
Delta compression using up to 8 threads
Compressing objects: 100% (21/21), done.
Writing objects: 100% (21/21), 9.11 KiB | 9.11 MiB/s, done.
Total 21 (delta 9), reused 0 (delta 0), pack-reused 0
To ssh://git.launchpad.net/ubuntu/+source/crmsh
 * [new tag] upload/4.2.0-3ubuntu1 -> upload/4.2.0-3ubuntu1

$ debdiff crmsh*.dsc | diffstat -l
changelog
control
patches/series
patches/ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch
patches/ubuntu/lp1895728-78ce77e4-Low-hb_report-Fix-collecting-of-binary-data-bsc-1166.patch
patches/ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch
patches/ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-should-complete-resource.patch
patches/ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch
patches/ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch

[rafaeldtinoco@groovy ubuntu]$ dput ubuntu crmsh_4.2.0-3ubuntu1_source.changes
Checking signature on .changes
Uploading to ubuntu (via ftp to upload.ubuntu.com):
  Uploading crmsh_4.2.0-3ubuntu1.dsc: done.
  Uploading crmsh_4.2.0-3ubuntu1.debian.tar.xz: done.
  Uploading crmsh_4.2.0-3ubuntu1_source.buildinfo: done.
  Uploading crmsh_4.2.0-3ubuntu1_source.changes: done.
Successfully uploaded packages.

review: Approve

[Christian Ehrhardt  (paelzer)]

merged
 crmsh | 4.2.0-3ubuntu1 | groovy | source, all

diff --git a/debian/changelog b/debian/changelog
index 858a74c..9b3ff70 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,15 @@
+crmsh (4.2.0-3ubuntu1) groovy; urgency=medium
+
+  * Fixes before release: (LP: #1895728)
+    - [997f62d6] Fix: hb_report: disable dump all tasks stack into dmesg
+    - [cce28254] Fix: crmsh.spec: using mktemp to create tmp file
+    - [e098387e] Fix: bootstrap: use csync2 '-f' option correctly
+    - [669f13b5] Fix: bootstrap: Change condition to add stonith-sbd resource
+    - [ca1f0a7f] Fix: ui_resource: refresh <Tab> should complete resource first
+    - [78ce77e4] Low: hb_report: Fix collecting of binary data
+
+ -- Rafael David Tinoco <[e-mail]>  Wed, 16 Sep 2020 03:41:38 +0000
+
 crmsh (4.2.0-3) unstable; urgency=medium
 
   [ Debian Janitor ]
diff --git a/debian/control b/debian/control
index 4f9a87e..24a2782 100644
--- a/debian/control
+++ b/debian/control
@@ -1,7 +1,8 @@
 Source: crmsh
 Section: admin
 Priority: optional
-Maintainer: Debian HA Maintainers <[e-mail]>
+Maintainer: Ubuntu Developers <[e-mail]>
+XSBC-Original-Maintainer: Debian HA Maintainers <[e-mail]>
 Uploaders:
  Richard B Winters <[e-mail]>,
  Adrian Vondendriesch <[e-mail]>,
diff --git a/debian/patches/series b/debian/patches/series
index 94acfdf..07859a4 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -11,3 +11,12 @@
 0015-Fix-testsuite-errors.patch
 0016-Fix-python2-calls.patch
 0017-Fix-python3.8-unittest-compatibility.patch
+#
+# ubuntu: post release fixes (LP: #1895728)
+#
+ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch
+ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch
+ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch
+ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch
+ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-should-complete-resource.patch
+ubuntu/lp1895728-78ce77e4-Low-hb_report-Fix-collecting-of-binary-data-bsc-1166.patch
diff --git a/debian/patches/ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch b/debian/patches/ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch
new file mode 100644
index 0000000..334a853
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-669f13b5-Fix-bootstrap-Change-condition-to-add-stonith-sbd-re.patch
@@ -0,0 +1,40 @@
+From 669f13b518847287d9994e8f3a00ffd876b376df Mon Sep 17 00:00:00 2001
+From: liangxin1300 <[e-mail]>
+Date: Wed, 18 Mar 2020 20:15:08 +0800
+Subject: [PATCH] Fix: bootstrap: Change condition to add stonith-sbd
+ resource(bsc#1166967)
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/669f13b518847287d9994e8f3a00ffd876b376df
+Bug: https://github.com/ClusterLabs/crmsh/pull/545
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ crmsh/bootstrap.py | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/crmsh/bootstrap.py b/crmsh/bootstrap.py
+index 36d5149f..db78b7dd 100644
+--- a/crmsh/bootstrap.py
++++ b/crmsh/bootstrap.py
+@@ -660,7 +660,7 @@ def init_cluster_local():
+ 
+     # for cluster join, diskless_sbd flag is set in join_cluster() if
+     # sbd is running on seed host
+-    if configured_sbd_device() or _context.diskless_sbd:
++    if (configured_sbd_device() and _context.sbd_device) or _context.diskless_sbd:
+         invoke("systemctl enable sbd.service")
+     else:
+         invoke("systemctl disable sbd.service")
+@@ -1496,7 +1496,7 @@ op_defaults op-options: timeout=600 record-pending=true
+ rsc_defaults rsc-options: resource-stickiness=1 migration-threshold=3
+ """)
+ 
+-    if configured_sbd_device():
++    if configured_sbd_device() and _context.sbd_device:
+         if not invoke("crm configure primitive stonith-sbd stonith:external/sbd pcmk_delay_max=30s"):
+             error("Can't create stonith-sbd primitive")
+         if not invoke("crm configure property stonith-enabled=true"):
+-- 
+2.27.0
+
diff --git a/debian/patches/ubuntu/lp1895728-78ce77e4-Low-hb_report-Fix-collecting-of-binary-data-bsc-1166.patch b/debian/patches/ubuntu/lp1895728-78ce77e4-Low-hb_report-Fix-collecting-of-binary-data-bsc-1166.patch
new file mode 100644
index 0000000..54206a3
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-78ce77e4-Low-hb_report-Fix-collecting-of-binary-data-bsc-1166.patch
@@ -0,0 +1,128 @@
+From 78ce77e47b195fc2ef4a21d5d83886b178bbc0e1 Mon Sep 17 00:00:00 2001
+From: Petr Pavlu <[e-mail]>
+Date: Mon, 23 Mar 2020 13:19:20 +0100
+Subject: [PATCH] Low: hb_report: Fix collecting of binary data (bsc#1166962)
+
+When creating a resulting archive in a slave collector ('hb_report
+__slave'), the code tried to decode output produced by the tar utility
+as a UTF-8 string (crmutils.get_stdout_stderr() -> to_ascii()). This
+corrupted any binary files in the archive, such as gzipped pengine
+files.
+
+The patch fixes the problem by obtaining the output from the tar utility
+as a bytes object and writing it in the repr() form on stdout of the
+slave collector. Similarly, the receiving side is updated to accept this
+format.
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/78ce77e47b195fc2ef4a21d5d83886b178bbc0e1
+Bug: https://github.com/ClusterLabs/crmsh/pull/610
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ crmsh/utils.py         |  8 ++++++--
+ hb_report/constants.py |  1 +
+ hb_report/hb_report.in |  7 ++-----
+ hb_report/utillib.py   | 17 ++++++++++-------
+ 4 files changed, 19 insertions(+), 14 deletions(-)
+
+diff --git a/crmsh/utils.py b/crmsh/utils.py
+index fe08859c..04805734 100644
+--- a/crmsh/utils.py
++++ b/crmsh/utils.py
+@@ -725,7 +725,7 @@ def pipe_cmd_nosudo(cmd):
+     return rc
+ 
+ 
+-def get_stdout(cmd, input_s=None, stderr_on=True, shell=True):
++def get_stdout(cmd, input_s=None, stderr_on=True, shell=True, raw=False):
+     '''
+     Run a cmd, return stdout output.
+     Optional input string "input_s".
+@@ -743,10 +743,12 @@ def get_stdout(cmd, input_s=None, stderr_on=True, shell=True):
+                             stdout=subprocess.PIPE,
+                             stderr=stderr)
+     stdout_data, stderr_data = proc.communicate(input_s)
++    if raw:
++        return proc.returncode, stdout_data
+     return proc.returncode, to_ascii(stdout_data).strip()
+ 
+ 
+-def get_stdout_stderr(cmd, input_s=None, shell=True):
++def get_stdout_stderr(cmd, input_s=None, shell=True, raw=False):
+     '''
+     Run a cmd, return (rc, stdout, stderr)
+     '''
+@@ -758,6 +760,8 @@ def get_stdout_stderr(cmd, input_s=None, shell=True):
+                             stdout=subprocess.PIPE,
+                             stderr=subprocess.PIPE)
+     stdout_data, stderr_data = proc.communicate(input_s)
++    if raw:
++        return proc.returncode, stdout_data, stderr_data
+     return proc.returncode, to_ascii(stdout_data).strip(), to_ascii(stderr_data).strip()
+ 
+ 
+diff --git a/hb_report/constants.py b/hb_report/constants.py
+index c2d3925f..2d5cb346 100644
+--- a/hb_report/constants.py
++++ b/hb_report/constants.py
+@@ -8,6 +8,7 @@ ARGOPTS_VALUE = "f:t:l:u:X:p:L:e:E:n:MSDZVsvhdQ"
+ B_CONF = None
+ CIB_DIR = None
+ COMPRESS = config.report.compress
++COMPRESS_DATA_FLAG = "COMPRESS HB_REPORT DATA:::"
+ COMPRESS_PROG = ""
+ COMPRESS_EXT = ""
+ CORES_DIRS = None
+diff --git a/hb_report/hb_report.in b/hb_report/hb_report.in
+index 49138668..4dc1eabd 100755
+--- a/hb_report/hb_report.in
++++ b/hb_report/hb_report.in
+@@ -280,13 +280,10 @@ def run():
+     #     problem description template, and prints final notes
+     #
+     if is_collector():
+-        import io
+-        sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
+-
+         utillib.collect_info()
+         cmd = r"cd %s/.. && tar -h -cf - %s" % (constants.WORKDIR, constants.WE)
+-        code, out, err = crmutils.get_stdout_stderr(cmd)
+-        print(out)
++        code, out, err = crmutils.get_stdout_stderr(cmd, raw=True)
++        print("{}{}".format(constants.COMPRESS_DATA_FLAG, out))
+     else:
+         p_list = []
+         p_list.append(multiprocessing.Process(target=utillib.analyze))
+diff --git a/hb_report/utillib.py b/hb_report/utillib.py
+index 9ca08991..0ee1ffa0 100644
+--- a/hb_report/utillib.py
++++ b/hb_report/utillib.py
+@@ -1548,14 +1548,17 @@ def start_slave_collector(node, arg_str):
+                     log_warning(err)
+                 break
+ 
++    compress_data = ""
++    for data in out.split('\n'):
++        if data.startswith(constants.COMPRESS_DATA_FLAG):
++            # hb_report data from collector
++            compress_data = data.lstrip(constants.COMPRESS_DATA_FLAG)
++        else:
++            # log data from collector
++            print(data)
++
+     cmd = r"(cd {} && tar xf -)".format(constants.WORKDIR)
+-    # typeof out here will always "str"
+-    # input_s of get_stdout will always need bytes
+-    # different situation depend on whether found pe file
+-    if out.startswith("b'"):
+-        crmutils.get_stdout(cmd, input_s=eval(out))
+-    else:
+-        crmutils.get_stdout(cmd, input_s=out.encode('utf-8'))
++    crmutils.get_stdout(cmd, input_s=eval(compress_data))
+ 
+ 
+ def str_to_bool(v):
+-- 
+2.27.0
+
diff --git a/debian/patches/ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch b/debian/patches/ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch
new file mode 100644
index 0000000..9ee9e44
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-997f62d6-Fix-hb_report-disable-dump-all-tasks-stack-into-dmes.patch
@@ -0,0 +1,67 @@
+From 997f62d6ac86b388f464642c1e7db8d5205d21e6 Mon Sep 17 00:00:00 2001
+From: liangxin1300 <[e-mail]>
+Date: Fri, 29 Nov 2019 10:14:28 +0800
+Subject: [PATCH] Fix: hb_report: disable dump all tasks stack into
+ dmesg(bsc#1158060)
+
+We ever added this line to collect all process back-traces for helping to analyze a ocfs2 cluster hang problem,
+but in some actual production environments, there are lots of processes in a cluster node,
+this dump operation will take too much time, even bring a short system hang.
+Then, we should delete this operation during a generic cluster information collection.
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/997f62d6ac86b388f464642c1e7db8d5205d21e6
+Bug: https://github.com/ClusterLabs/crmsh/pull/486
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ hb_report/utillib.py | 23 +++++++++++++++++++++--
+ 1 file changed, 21 insertions(+), 2 deletions(-)
+
+diff --git a/hb_report/utillib.py b/hb_report/utillib.py
+index 69383717..9ca08991 100644
+--- a/hb_report/utillib.py
++++ b/hb_report/utillib.py
+@@ -1649,11 +1649,29 @@ def time_status():
+     time_f = os.path.join(constants.WORKDIR, constants.TIME_F)
+     crmutils.str2file(out_string, time_f)
+ 
++
++def dump_D_process():
++    '''
++    dump D-state process stack
++    '''
++    out_string = ""
++    _, out, _ = crmutils.get_stdout_stderr("ps aux|awk '$8 ~ /^D/{print $2}'")
++    len_D_process = len(out.split('\n')) if out else 0
++    out_string += "Dump D-state process stack: {}\n".format(len_D_process)
++    if len_D_process == 0:
++        return out_string
++    for pid in out.split('\n'):
++        _, cmd_out, _ = crmutils.get_stdout_stderr("cat /proc/{}/comm".format(pid))
++        out_string += "pid: {}     comm: {}\n".format(pid, cmd_out)
++        _, stack_out, _ = crmutils.get_stdout_stderr("cat /proc/{}/stack".format(pid))
++        out_string += stack_out + "\n\n"
++    return out_string
++
++
+ def dump_ocfs2():
+     ocfs2_f = os.path.join(constants.WORKDIR, constants.OCFS2_F)
+     with open(ocfs2_f, "w") as f:
+-        #dump all tasks stack into dmesg
+-        os.system("echo t > /proc/sysrq-trigger")
++        f.write(dump_D_process())
+ 
+         cmds = [ "dmesg",  "ps -efL", "lsof",
+                 "lsblk -o 'NAME,KNAME,MAJ:MIN,FSTYPE,LABEL,RO,RM,MODEL,SIZE,OWNER,GROUP,MODE,ALIGNMENT,MIN-IO,OPT-IO,PHY-SEC,LOG-SEC,ROTA,SCHED,MOUNTPOINT'",
+@@ -1670,6 +1688,7 @@ def dump_ocfs2():
+             f.write("# %s\n"%(cmd))
+             f.write(out)
+ 
++
+ def touch_dc():
+     if constants.SKIP_LVL:
+         return
+-- 
+2.27.0
+
diff --git a/debian/patches/ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-should-complete-resource.patch b/debian/patches/ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-should-complete-resource.patch
new file mode 100644
index 0000000..a65a61d
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-ca1f0a7f-Fix-ui_resource-refresh-Tab-should-complete-resource.patch
@@ -0,0 +1,31 @@
+From ca1f0a7fb5c424a406b7f85dd9903c4508f0c963 Mon Sep 17 00:00:00 2001
+From: liangxin1300 <[e-mail]>
+Date: Sun, 29 Mar 2020 19:48:38 +0800
+Subject: [PATCH] Fix: ui_resource: refresh <Tab> should complete resource
+ first(bsc#1167220)
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/ca1f0a7fb5c424a406b7f85dd9903c4508f0c963
+Bug: https://bugzilla.opensuse.org/show_bug.cgi?id=1167220
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ crmsh/ui_resource.py | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/crmsh/ui_resource.py b/crmsh/ui_resource.py
+index f8dee734..e52797a3 100644
+--- a/crmsh/ui_resource.py
++++ b/crmsh/ui_resource.py
+@@ -604,7 +604,7 @@ class RscMgmt(command.UI):
+                                     rsc, cmd, attr, value)
+ 
+     @command.alias('reprobe')
+-    @command.completers(compl.nodes)
++    @command.completers(compl.resources, compl.nodes)
+     def do_refresh(self, context, rsc=None, node=None, force=False):
+         'usage: refresh [<rsc>] [<node>] [force]'
+         return self._refresh_cleanup("refresh", rsc, node, force)
+-- 
+2.27.0
+
diff --git a/debian/patches/ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch b/debian/patches/ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch
new file mode 100644
index 0000000..b3aaa5f
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-cce28254-Fix-crmsh.spec-using-mktemp-to-create-tmp-file-bsc-1.patch
@@ -0,0 +1,30 @@
+From cce28254b3b17c9a4ecdfce1012fb41f1f1abc78 Mon Sep 17 00:00:00 2001
+From: liangxin1300 <[e-mail]>
+Date: Tue, 7 Jan 2020 09:25:19 +0800
+Subject: [PATCH] Fix: crmsh.spec: using mktemp to create tmp file(bsc#1154163)
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/cce28254b3b17c9a4ecdfce1012fb41f1f1abc78
+Bug: https://bugzilla.opensuse.org/show_bug.cgi?id=1154163
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ crmsh.spec.in | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/crmsh.spec.in b/crmsh.spec.in
+index f6553b33..18cf5b1f 100644
+--- a/crmsh.spec.in
++++ b/crmsh.spec.in
+@@ -201,7 +201,7 @@ fi
+ # Run regression tests after installing the package
+ # NB: this is called twice by OBS, that's why we touch the file
+ %post test
+-testfile=/tmp/.crmsh_regression_tests_ran
++testfile=`mktemp -t .crmsh_regression_tests_ran_XXXXXX`
+ # check if time in file is less than 2 minutes ago
+ if [ -e $testfile ] && [ "$(( $(date +%s) - $(cat $testfile) ))" -lt 120 ]; then
+ 	echo "Skipping regression tests..."
+-- 
+2.27.0
+
diff --git a/debian/patches/ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch b/debian/patches/ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch
new file mode 100644
index 0000000..3bcfcc7
--- /dev/null
+++ b/debian/patches/ubuntu/lp1895728-e098387e-Fix-bootstrap-use-csync2-f-option-correctly-bsc-1166.patch
@@ -0,0 +1,51 @@
+From e098387e2ae79893cfff3ad91d5a6e26d62ecfc8 Mon Sep 17 00:00:00 2001
+From: liangxin1300 <[e-mail]>
+Date: Sun, 15 Mar 2020 09:27:44 +0800
+Subject: [PATCH] Fix: bootstrap: use csync2 '-f' option correctly(bsc#1166684)
+
+Origin: upstream, https://github.com/ClusterLabs/crmsh/commit/e098387e2ae79893cfff3ad91d5a6e26d62ecfc8
+Bug: https://github.com/ClusterLabs/crmsh/pull/542
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1895728
+Reviewed-by: Rafael David Tinoco <[e-mail]>
+Last-Update: 2020-09-15
+---
+ crmsh/bootstrap.py | 16 ++++++++++++----
+ 1 file changed, 12 insertions(+), 4 deletions(-)
+
+diff --git a/crmsh/bootstrap.py b/crmsh/bootstrap.py
+index e8e00cda..14e61066 100644
+--- a/crmsh/bootstrap.py
++++ b/crmsh/bootstrap.py
+@@ -784,9 +784,17 @@ include /etc/pacemaker/authkey;
+ 
+ 
+ def csync2_update(path):
+-    invoke("csync2 -rm %s" % (path))
+-    invoke("csync2 -rf %s" % (path))
+-    invoke("csync2 -rxv %s" % (path))
++    '''
++    Sync path to all peers
++
++    If there was a conflict, use '-f' to force this side to win
++    '''
++    invoke("csync2 -rm {}".format(path))
++    if invoke("csync2 -rxv {}".format(path)):
++        return
++    invoke("csync2 -rf {}".format(path))
++    if not invoke("csync2 -rxv {}".format(path)):
++        warn("{} was not synced".format(path))
+ 
+ 
+ def init_csync2_remote():
+@@ -1740,7 +1748,7 @@ def join_csync2(seed_host):
+     # they haven't gone to all nodes in the cluster, which means a
+     # subseqent join of another node can fail its sync of corosync.conf
+     # when it updates expected_votes.  Grrr...
+-    if not invoke('ssh -o StrictHostKeyChecking=no root@%s "csync2 -mr / ; csync2 -fr / ; csync2 -xv"' % (seed_host)):
++    if not invoke('ssh -o StrictHostKeyChecking=no root@{} "csync2 -rm /; csync2 -rxv || csync2 -rf / && csync2 -rxv"'.format(seed_host)):
+         print("")
+         warn("csync2 run failed - some files may not be sync'd")
+ 
+-- 
+2.27.0
+