[Bug testsuite/16326] New: client.exp and local stap-server interactions

[Bug testsuite/16326] New: client.exp and local stap-server interactions

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

            Bug ID: 16326
           Summary: client.exp and local stap-server interactions
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: testsuite
          Assignee: systemtap at sourceware dot org
          Reporter: mcermak at redhat dot com

Created attachment 7321
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7321&action=edit
proposed patch

client.exp doesn't work fine if local stap-server runs. If valid trust for it
exists, testcase fails like this:

FAIL: No new signing servers
FAIL: No longer trusted after revokation by host name
XPASS: New signing servers
FAIL: No longer trusted as a signer after revokation

If local stap-server is stopped but valid trust for it exists, then the
testcase result is:

FAIL: Verify existing online server list
FAIL: Verify all existing server list

These are duplicates (PR16325) and arise because `stap --list-servers=all`
returns one record (with address=offline port=unknown) and so
existing_online_servers != all_existing_servers.

So I guess it's better to stop local stap-server if it runs and revoke it's
trust at our testcase startup.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

--- Comment #1 from mcermak <mcermak at redhat dot com> ---
Also foreign stap-servers randomly running around break the testcase if theyŕe
not trusted. The testcase aparently somehow expect some such servers to run:

[ ... ]
# There may be other servers running. Let's keep track of
# them.
#
[ ... ]

But if the're not trusted, the testcase simply fails:

FAIL: Verify existing trusted server list
FAIL: Verify existing signing server list

It's because `stap --list-servers=trusted` differs from `stap
--list-servers=all` and `stap --list-servers=signing` differs from `stap
--list-servers=all`. 

Described behaviour may complicate automated usage of the testcase. One
possible solution would be to automatically add trust fo all such servers.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[brolley at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Dave Brolley <brolley at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |brolley at redhat dot com

--- Comment #2 from Dave Brolley <brolley at redhat dot com> ---
The client.exp tests were originally designed to tolerate local and remote
servers which are already running. If these servers are not tolerated, then the
test should be fixed.

Thanks for the bug report!

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Jonathan Lebon <jlebon at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jlebon at redhat dot com

--- Comment #3 from Jonathan Lebon <jlebon at redhat dot com> ---
Created attachment 7326
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7326&action=edit
fix client.exp

This patch restores awareness for other possible servers running and fixes the
failures mentioned above. It also significantly simplifies the code by
introducing a few array utility procedures.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Jonathan Lebon <jlebon at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #4 from Jonathan Lebon <jlebon at redhat dot com> ---
Merged into master. Commit 8b68fe5.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

--- Comment #5 from Jonathan Lebon <jlebon at redhat dot com> ---
*** Bug 16325 has been marked as a duplicate of this bug. ***

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

--- Comment #6 from Martin Cermak <mcermak at redhat dot com> ---
Created attachment 7329
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7329&action=edit
logfile acquired using client.exp 8b68fe58c64f

Not yet there. The testcase fails in case there is a local trusted server
*running*.

Have one local and one remote servers running before executing our testcase.
Now edit the testcase like this:

verbose -log "SLEEP..."
exec sleep 40

# Now start our own server and make sure we can work with it.
if {! [setup_server] || $avahi_ok_p != 1} {
    untested "Compile-server client tests against a server"
    return;
}

verbose -log "SLEEP..."
exec sleep 40

1) check before running the testcase:

# stap --list-servers=all; stap --list-servers=online; stap
--list-servers=signer; stap --list-servers=trusted
Systemtap Compile Server Status for 'all'
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
Systemtap Compile Server Status for 'online'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'signer'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'trusted'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"

2) check within the first sleep

# stap --list-servers=all; stap --list-servers=online; stap
--list-servers=signer; stap --list-servers=trusted
Systemtap Compile Server Status for 'all'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'online'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'signer'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'trusted'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"

3) check within the second sleep

# stap --list-servers=all; stap --list-servers=online; stap
--list-servers=signer; stap --list-servers=trusted
Systemtap Compile Server Status for 'all'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=60632
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b9:55:00"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1 address=YY.YY.YY.YY port=48702 sysinfo="3.10.0-55.el7.x86_64 x86_64"
version=2.4 certinfo="00:9d:a3:a2:73"
 host=box1.blah.redhat.com address=offline port=unknown sysinfo="unknown"
version=unknown certinfo="00:9d:a3:a2:73"
Systemtap Compile Server Status for 'online'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=60632
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b9:55:00"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1 address=YY.YY.YY.YY port=48702 sysinfo="3.10.0-55.el7.x86_64 x86_64"
version=2.4 certinfo="00:9d:a3:a2:73"
Systemtap Compile Server Status for 'signer'
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1 address=YY.YY.YY.YY port=48702 sysinfo="3.10.0-55.el7.x86_64 x86_64"
version=2.4 certinfo="00:9d:a3:a2:73"
 host=box1.blah.redhat.com address=offline port=unknown sysinfo="unknown"
version=unknown certinfo="00:9d:a3:a2:73"
Systemtap Compile Server Status for 'trusted'
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1 address=YY.YY.YY.YY port=48702 sysinfo="3.10.0-55.el7.x86_64 x86_64"
version=2.4 certinfo="00:9d:a3:a2:73"
 host=box1.blah.redhat.com address=offline port=unknown sysinfo="unknown"
version=unknown certinfo="00:9d:a3:a2:73"

4) check after the testcasse finish

# stap --list-servers=all; stap --list-servers=online; stap
--list-servers=signer; stap --list-servers=trusted
Systemtap Compile Server Status for 'all'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'online'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'signer'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
Systemtap Compile Server Status for 'trusted'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"

I see 3 issues:

1) After *careful* text-diffing these outputs it appears that one can't rely on
the line order produced by --list-servers, which may complicate array key-val
comparisons in the testcase.

2) I'm a bit confused about what happens when additional local server is
started: Let's have following initial state:

# stap --list-servers=all; netstat -tlpn | grep stap
Systemtap Compile Server Status for 'all'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=48702
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:a3:a2:73"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
tcp6       0      0 :::48702                :::*                    LISTEN     
27016/stap-serverd

Now let's fire up an extra local server by executing
`/usr/libexec/systemtap/stap-serv --log=/tmp/mystaplog` under some regular user
(this is what the testcase does). Now I got:

# stap --list-servers=all; netstat -tlpn | grep stap
Systemtap Compile Server Status for 'all'
 host=box1.blah.redhat.com address=YY.YY.YY.YY port=35053
sysinfo="3.10.0-55.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b9:70:b9"
 host=box3.blah.redhat.com address=XX.XX.XX.XX port=56334
sysinfo="3.10.0-61.el7.x86_64 x86_64" version=2.4 certinfo="00:9d:b8:18:bb"
 host=box1 address=YY.YY.YY.YY port=48702 sysinfo="3.10.0-55.el7.x86_64 x86_64"
version=2.4 certinfo="00:9d:a3:a2:73"
 host=box1.blah.redhat.com address=offline port=unknown sysinfo="unknown"
version=unknown certinfo="00:9d:a3:a2:73"
tcp6       0      0 :::35053                :::*                    LISTEN     
17991/stap-serverd
tcp6       0      0 :::48702                :::*                    LISTEN     
27016/stap-serverd

Not sure what does "host=box1.blah.redhat.com address=offline" mean..

3) It might be useful to move those newly created array_* procs to some shared
library, possibly lib/systemtap.exp

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Jonathan Lebon <jlebon at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Depends on|                            |16342

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Jonathan Lebon <jlebon at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |---

--- Comment #7 from Jonathan Lebon <jlebon at redhat dot com> ---
Reopened until concerns in comment #6 are addressed.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Bug 16326 depends on bug 16342, which changed state.

Bug 16342 Summary: stap --list-servers lists spurious line when having two stap-serverd instances
https://sourceware.org/bugzilla/show_bug.cgi?id=16342

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

--- Comment #8 from Jonathan Lebon <jlebon at redhat dot com> ---
(In reply to Martin Cermak from comment #6)
> Created attachment 7329 [details]
> logfile acquired using client.exp 8b68fe58c64f
> 
> Not yet there. The testcase fails in case there is a local trusted server
> *running*.

Martin, can you verify that the testcase passes now?

Thanks!

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

--- Comment #9 from Martin Cermak <mcermak at redhat dot com> ---
One detail:
(In reply to Martin Cermak from comment #6)
> 3) It might be useful to move those newly created array_* procs to some
> shared library, possibly lib/systemtap.exp
It's just that I think it might be useful sometime later.

Otherwise everything looks pretty good to me now! Thanks for your fixes,
Jonathan!

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug testsuite/16326] client.exp and local stap-server interactions

[jlebon at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16326

Jonathan Lebon <jlebon at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #10 from Jonathan Lebon <jlebon at redhat dot com> ---
(In reply to Martin Cermak from comment #9)
> One detail:
> (In reply to Martin Cermak from comment #6)
> > 3) It might be useful to move those newly created array_* procs to some
> > shared library, possibly lib/systemtap.exp
> It's just that I think it might be useful sometime later.

I think we'll leave them there until another test requires them. No other test
for now uses arrays that way.

> Otherwise everything looks pretty good to me now! Thanks for your fixes,
> Jonathan!

Thanks for testing Martin!

-- 
You are receiving this mail because:
You are the assignee for the bug.