openSUSE Project Management Tool: Issueshttps://progress.opensuse.org/https://progress.opensuse.org/themes/openSUSE/favicon/favicon.ico?15829177842021-11-29T13:49:37ZopenSUSE Project Management Tool
Redmine openQA Project - action #103203 (Resolved): openqa.opensuse.org consistently fails to boot from v...https://progress.opensuse.org/issues/1032032021-11-29T13:49:37Zdancermakdcermak@suse.com
<a name="Observation"></a>
<h2 >Observation<a href="#Observation" class="wiki-anchor">¶</a></h2>
<p>We are facing some weird issues with vmdk disk images: while openQA can boot from these without issues on my local development machine, it consistently fails to do so on o3. E.g.: <a href="https://openqa.opensuse.org/tests/2062448" class="external">https://openqa.opensuse.org/tests/2062448</a>, <a href="https://openqa.opensuse.org/tests/2062446" class="external">https://openqa.opensuse.org/tests/2062446</a>. This issue appears to be also intermingled with asset download failures on o3, which troubled the vmdk disk image group, but this did not appear anymore in the last run.</p>
<a name="Steps-to-reproduce-locally"></a>
<h2 >Steps to reproduce locally<a href="#Steps-to-reproduce-locally" class="wiki-anchor">¶</a></h2>
<ul>
<li><code>git clone https://github.com/OSInside/kiwi-functional-tests</code></li>
<li><code>poetry install</code></li>
<li><code>poetry run ./settings.py --server $MY_INSTANCE --server-scheme http</code></li>
</ul>
<p>and then clone one of the failing jobs or schedule a full test run (this will schedule <strong>many</strong> jobs):</p>
<ul>
<li><code>poetry run ./schedule_test_run.py --server $MY_INSTANCE --server-scheme http -vd Tumbleweed+opensuse --use-https-for-asset-download</code></li>
</ul>
<a name="Steps-to-reproduce-remotely"></a>
<h2 >Steps to reproduce remotely<a href="#Steps-to-reproduce-remotely" class="wiki-anchor">¶</a></h2>
<ul>
<li>clone the failed openQA job on o3 directly</li>
</ul>
<a name="Problem"></a>
<h2 >Problem<a href="#Problem" class="wiki-anchor">¶</a></h2>
<p>openQA appears to be unable to boot from the vmdk disk images: they are not considered to be bootable devices by the BIOS/EFI firmware (this is independent of whether we are booting from BIOS or EFI).</p>
<a name="Suggestion"></a>
<h2 >Suggestion<a href="#Suggestion" class="wiki-anchor">¶</a></h2>
<p>Sadly, none.</p>
<a name="Workaround"></a>
<h2 >Workaround<a href="#Workaround" class="wiki-anchor">¶</a></h2>
<p>None known.</p>
openQA Project - action #94735 (Resolved): needles not found in `needles` subdirectory when CASED...https://progress.opensuse.org/issues/947352021-06-25T13:23:34Zdancermakdcermak@suse.com
<a name="Observation"></a>
<h2 >Observation<a href="#Observation" class="wiki-anchor">¶</a></h2>
<p>openQA does not appear to take the <code>needles</code> subdirectory into account for loading needles when <code>CASEDIR</code> is a git repository.</p>
<a name="Steps-to-reproduce"></a>
<h2 >Steps to reproduce<a href="#Steps-to-reproduce" class="wiki-anchor">¶</a></h2>
<ul>
<li>Clone a job and set <code>CASEDIR</code> to a git repository with a <code>needles</code> subdirectory in that git repository</li>
<li>If the <code>needles</code> subdirectory contains needles not present on the instance, then openQA will not find them.</li>
</ul>
<a name="Workaround"></a>
<h2 >Workaround<a href="#Workaround" class="wiki-anchor">¶</a></h2>
<p>Setting <code>NEEDLES_DIR</code> to <code>CASEDIR</code> works around that problem. Setting <code>NEEDLES_DIR</code> to <code>needles/</code> on the other hand does not work.</p>
<p>Example failure: <a href="https://openqa.opensuse.org/tests/1806254" class="external">https://openqa.opensuse.org/tests/1806254</a> (-> failed)</p>
<p>cloned with <code>NEEDLES_DIR='needles/'</code>: <a href="https://openqa.opensuse.org/tests/1806460" class="external">https://openqa.opensuse.org/tests/1806460</a> (-> failed)</p>
<p>cloned with <code>NEEDLES_DIR='https://github.com/dcermak/kiwi-functional-tests#basic_functional_test'</code>: <a href="https://openqa.opensuse.org/tests/1806588" class="external">https://openqa.opensuse.org/tests/1806588</a> (-> passed)</p>
<a name="Suggestions"></a>
<h2 >Suggestions<a href="#Suggestions" class="wiki-anchor">¶</a></h2>
<ul>
<li>Be careful (see comment <a href="https://progress.opensuse.org/issues/94735#note-3" class="external">https://progress.opensuse.org/issues/94735#note-3</a>)</li>
</ul>
openQA Project - action #94084 (Rejected): POST /job_templates_scheduling fails silently when the...https://progress.opensuse.org/issues/940842021-06-16T08:25:16Zdancermakdcermak@suse.com
<p>I have started out with a completely empty openQA instance and tried to apply a job template via a script that uses the /job_templates_scheduling route to set the job template. The template includes the following default:</p>
<pre><code class="yaml syntaxhl" data-language="yaml"><span class="na">defaults</span><span class="pi">:</span>
<span class="na">x86_64</span><span class="pi">:</span>
<span class="na">machine</span><span class="pi">:</span> <span class="s">64bit</span>
<span class="na">priority</span><span class="pi">:</span> <span class="m">50</span>
</code></pre>
<p>But as the instance is completely fresh, it does not have the 64 bit machine defined and the above template fails to apply.</p>
<p>Unfortunately, the API swallows this failure and completely and still responds with a 204. The failure only becomes apparent when checking the logs of the webui, which includes:</p>
<pre><code>webui_1 | [error] Machine '64bit' is invalid
webui_2 | [error] Machine '64bit' is invalid
webui_1 | [error] Machine '64bit' is invalid
webui_1 | [error] Machine '64bit' is invalid
</code></pre> openQA Project - action #91965 (Workable): Enable switching to the qemu serial0 linehttps://progress.opensuse.org/issues/919652021-04-29T19:13:21Zdancermakdcermak@suse.com
<a name="Motivation"></a>
<h2 >Motivation<a href="#Motivation" class="wiki-anchor">¶</a></h2>
<p>I am writing a test for an appliance that launches an installer after confirmation (send enter) over the serial line. This can be achieved rather easily in virt-manager by switching the default console from "graphical spice console" to "serial 1" and pressing enter there. Unfortunately, I cannot achieve the same in openQA. It records the output from the serial line into <code>/var/lib/openqa/pool/$worker_id/serial0</code>, but I am not able to switch to this serial line or send something to it.</p>
<p>Is this somehow possible or generally unsupported?</p>
<a name="Acceptance-criteria"></a>
<h2 >Acceptance criteria<a href="#Acceptance-criteria" class="wiki-anchor">¶</a></h2>
<ul>
<li><strong>AC1:</strong> The primary serial console in a qemu SUT can be written to from openQA tests</li>
</ul>
<a name="Suggestions"></a>
<h2 >Suggestions<a href="#Suggestions" class="wiki-anchor">¶</a></h2>
<p>Right now we hardcode "serial0" as the first and only device that is setup as a serial terminal for a qemu instance. This device is used by writing command outputs to a logfile by forwarding script output to the variable "$SERIALDEV", e.g. ttyS0 in GNU/Linux in script commands like <code>assert_script_run 'true'</code> that end up being executed in SUTs like <code>true ; echo ...$? > /dev/ttyS0</code>. The logfile is then read by commands like <code>wait_serial</code>. Maybe we can make the name of the device+logfile configurable and add an option to configure a serial device before that internal logging device</p>
openQA Tests - action #77074 (Rejected): [qe-core] load_extra_tests_desktop is unusedhttps://progress.opensuse.org/issues/770742020-11-06T11:33:21Zdancermakdcermak@suse.com
<p>The function <code>load_extra_tests_desktop</code> from <code>lib/main_common.pm</code> line 1484 onward, appears to be unused. Can it be dropped?</p>
openQA Tests - action #77071 (Workable): [qe-core] Please move the declarative/yaml test flow sch...https://progress.opensuse.org/issues/770712020-11-06T09:31:14Zdancermakdcermak@suse.com
<p>The declarative YAML scheduler is a pretty useful thing to have, but it is currently only available in the os-autoinst-distri-opensuse repository. Would it be possible to move it to the upstream test api, so that it can be used by other test users as well?</p>
<a name="Suggestions"></a>
<h2 >Suggestions<a href="#Suggestions" class="wiki-anchor">¶</a></h2>
<p>Start by giving a look at: <a href="https://github.com/os-autoinst/os-autoinst-distri-opensuse/blob/master/lib/scheduler.pm" class="external">https://github.com/os-autoinst/os-autoinst-distri-opensuse/blob/master/lib/scheduler.pm</a></p>
openQA Project - action #70333 (Resolved): providing a git branch for NEEDLES_DIR does not appear...https://progress.opensuse.org/issues/703332020-08-20T15:52:21Zdancermakdcermak@suse.com
<p>I have tried to launch a test with a custom set of needles from a git branch on github by using the <code>NEEDLES_DIR</code> variable. Unfortunately, this does not appear to work reliably. For example these two tests: <a href="https://openqa.opensuse.org/tests/1367310" class="external">https://openqa.opensuse.org/tests/1367310</a> and <a href="https://openqa.opensuse.org/tests/1367262" class="external">https://openqa.opensuse.org/tests/1367262</a> appear to clone the git branch, but then the test fails as it cannot find one of the new needles. The same thing happened for a completely unrelated test as well: <a href="https://openqa.opensuse.org/tests/1365359" class="external">https://openqa.opensuse.org/tests/1365359</a>.</p>
openQA Project - action #69490 (Resolved): openqa webui sometimes displays the test suite instead...https://progress.opensuse.org/issues/694902020-07-31T16:53:01Zdancermakdcermak@suse.com
<p>During testing I have frequently had the issue that the matching needle drop down menu would no longer display the matched needles but instead a part of the test suite page (see the attached screenshot). The issue disappears when I reload the page.</p>
<p>This is my local openQA instance: openQA-4.6.1595658530.ccadcf580-1.1.noarch</p>
openQA Project - action #64361 (Resolved): load_templates or something else creates invalid test ...https://progress.opensuse.org/issues/643612020-03-10T14:17:58Zdancermakdcermak@suse.com
<p>I have setup a new openQA test instance on a Leap 15.1 VM to try and reproduce some test failures on o3. However, during the setup, I found that (probably) the <code>load_templates</code> script produced odd test suites. For instance:</p>
<pre><code>EXTRA=ARRAY(0x5618f1aa7aa8)
SYSTEM_ROLE=ARRAY(0x5618f1af9c18)
</code></pre>
<p>for <code>MicroOS-admin</code> instead of the expected:</p>
<pre><code>EXTRA=FEATURES
SYSTEM_ROLE=admin
</code></pre>
<p>as it shows up on o3.</p>
<p>This appears to affect <strong>all</strong> settings of every test suite that was added. The variable names are correct, as are the test names and the descriptions, just the values are wrong.</p>
<p>For reference, I have populated the settings as follows:</p>
<pre><code>/usr/share/openqa/script/fetchneedles
/var/lib/openqa/share/tests/opensuse/products/opensuse/templates --host openqa.opensuse.org --apikey $APIKEY --apisecret $APISECRET
/usr/share/openqa/script/dump_templates --host openqa.opensuse.org > templates.json
/usr/share/openqa/script/load_templates --host localhost --update templates.json
</code></pre> openQA Tests - action #62864 (Resolved): [qe-core][opensuse][functional] problem_detection relies...https://progress.opensuse.org/issues/628642020-01-31T14:37:21Zdancermakdcermak@suse.com
<p>I have tried developing a openQA test and saw that the function <code>problem_detection()</code> is invoked after a test failure. In the resulting video I saw that it invokes <code>coredumpctl</code>, however <code>coredumpctl</code> was not installed (at least in the <code>opensuse-Tumbleweed-x86_64-20200129-gnome-x11@64bit.qcow2</code> disk image).</p>
<p>Should the function try to install it? Or maybe we need to install <code>systemd-coredump</code> by default in the images that use <code>problem_detection()</code>.</p>
<a name="Tasks"></a>
<h2 >Tasks<a href="#Tasks" class="wiki-anchor">¶</a></h2>
<ol>
<li>Implement a check in <code>problem_detection()</code> to know if coredumpctl is installed, and if not, install it.</li>
</ol>
openQA Project - action #62510 (Resolved): openqa-bootstrap-container does not install qemu-tools...https://progress.opensuse.org/issues/625102020-01-22T05:01:26Zdancermakdcermak@suse.com
<p>The <code>openqa-bootstrap-container</code> script does not install the packages:</p>
<ul>
<li>qemu-tools</li>
<li>perl-LWP-Protocol-https</li>
</ul>
<p>inside the container which leaves the test instance pretty crippled, as it:</p>
<ul>
<li>cannot download assets via https</li>
<li>cannot launch a worker as they die with qemu errors</li>
</ul>
openSUSE admin - tickets #61826 (Closed): Icon urls returned by the /distributions API route on O...https://progress.opensuse.org/issues/618262020-01-07T09:06:54Zdancermakdcermak@suse.com
<p>There appears to be an issue on OBS: if you request the list of distributions via the API:</p>
<pre><code>curl --user $OBS_USERNAME:$OBS_PASSWORD 'https://api.opensuse.org/distributions'
</code></pre>
<p>then you'll be presented with a bunch of distros with urls to their icons. However, all the urls that I tried are dead links.</p>
openQA Project - action #60118 (Resolved): load_templates does not load JobGroups (unless they ex...https://progress.opensuse.org/issues/601182019-11-21T14:06:20Zdancermakdcermak@suse.com
<p>I have tried to setup a local openQA instance in a container as described in the official documentation and after getting all the tests (<a href="http://open.qa/docs/#_getting_tests">http://open.qa/docs/#_getting_tests</a>), I tried to fetch the job groups, templates, etc from o3. Unfortunately using dump_templates & load_templates fails:</p>
<pre><code>openqa1:/ # /usr/share/openqa/script/dump_templates --host openqa.opensuse.org > templates.json
openqa1:/ # /usr/share/openqa/script/load_templates --host localhost templates.json
ERROR: 400 - Bad Request
{ error => ["Job group OBS Unstable not found"], error_status => 400 }
ERROR: 400 - Bad Request
{
error => [{ message => "Missing property.", path => "/scenarios" }],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group openSUSE Leap 15.0 Incidents not found"],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group Staging Projects not found"],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group openSUSE Leap 42.3 Updates not found"],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group openSUSE Leap Staging Projects not found"],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group openSUSE Leap 15.1 AArch64 Images not found"],
error_status => 400,
}
ERROR: 400 - Bad Request
{
error => ["Job group openSUSE Leap 15.1 Updates not found"],
error_status => 400,
}
SNIP
ERROR: 414 - Request-URI Too Long
"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\"\n \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd\">\n<html xmlns=\"http://www.w3.
org/1999/xhtml\" lang=\"en\" xml:lang=\"en\">\n<head>\n<title>Submitted URI too large!</title>\n<link rev=\"made\" href=\"mailto:%5bno%20address%20given%5d\" />\n<style type=\"text/css\"><!--
/*--><![CDATA[/*><!--*/ \n body { color: #000000; background-color: #FFFFFF; }\n a:link { color: #0000CC; }\n p, address {margin-left: 3em;}\n span {font-size: smaller;}\n/*]]>*/-
-></style>\n</head>\n\n<body>\n<h1>Submitted URI too large!</h1>\n<p>\n\n\n The length of the requested URL exceeds the capacity limit for\n\tthis server. The request cannot be processed.\
n \n</p>\n<p>\nIf you think this is a server error, please contact\nthe <a href=\"mailto:%5bno%20address%20given%5d\">webmaster</a>.\n\n</p>\n\n<h2>Error 414</h2>\n<address>\n <a href=\"/\
">openqa1</a><br />\n <span>Apache</span>\n</address>\n</body>\n</html>\n\n"
SNIP
{
JobGroups => { added => 0, of => 40 },
JobTemplates => { added => 0, of => 0 },
Machines => { added => 49, of => 49 },
Products => { added => 140, of => 140 },
TestSuites => { added => 331, of => 331 },
}
</code></pre>
<p>I have attached the templates.json file that was produced by dump_templates.</p>
openSUSE admin - tickets #57530 (Resolved): Redmine cannot handle UTF8 characters in the lastnamehttps://progress.opensuse.org/issues/575302019-09-30T11:25:19Zdancermakdcermak@suse.com
<p>I can only login to progress.opensuse.org when I convert my lastname from "Čermák" to "Cermak", otherwise redmine dies with (<a href="https://paste.opensuse.org/b4d3c7a7):" class="external">https://paste.opensuse.org/b4d3c7a7):</a></p>
<p>Started GET "/login" for 89.12.147.145 at Mon Sep 30 08:52:01 +0000 2019<br>
Processing by AccountController#login as HTML<br>
Current user: anonymous<br>
Completed 500 Internal Server Error in 336.5ms</p>
<p>ActiveRecord::StatementInvalid (Mysql2::Error: Incorrect string value: '\xC4\x8Cerm\xC3...' for column 'lastname' at row 1: UPDATE <code>users</code> SET <code>login</code> = 'dancermak', <code>updated_on</code> = '2019-09-30 08:52:01', <code>lastname</code> = 'Čermák'<br>
WHERE <code>users</code>.<code>type</code> IN ('User', 'AnonymousUser') AND <code>users</code>.<code>id</code> = 33662):<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract_mysql_adapter.rb:245:in <code>query'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract_mysql_adapter.rb:245:in</code>execute'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract_adapter.rb:280:in <code>log'<br>
activesupport (3.2.17) lib/active_support/notifications/instrumenter.rb:20:in</code>instrument'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract_adapter.rb:275:in <code>log'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract_mysql_adapter.rb:245:in</code>execute'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/mysql2_adapter.rb:213:in <code>execute'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/mysql2_adapter.rb:240:in</code>exec_update'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract/database_statements.rb:96:in <code>update'<br>
activerecord (3.2.17) lib/active_record/connection_adapters/abstract/query_cache.rb:14:in</code>update'<br>
activerecord (3.2.17) lib/active_record/persistence.rb:359:in <code>update'<br>
activerecord (3.2.17) lib/active_record/locking/optimistic.rb:68:in</code>update'<br>
activerecord (3.2.17) lib/active_record/attribute_methods/dirty.rb:74:in <code>update'<br>
activerecord (3.2.17) lib/active_record/timestamp.rb:71:in</code>update'<br>
activerecord (3.2.17) lib/active_record/callbacks.rb:272:in <code>update'<br>
activesupport (3.2.17) lib/active_support/callbacks.rb:403:in</code><u>run</u><u>564565134</u><u>update</u><u>4</u><u>callbacks'<br>
activesupport (3.2.17) lib/active_support/callbacks.rb:405:in <code>send'<br>
activesupport (3.2.17) lib/active_support/callbacks.rb:405:in</code></u>_run_callback'<br>
activesupport (3.2.17) lib/active_support/callbacks.rb:385:in `_run_update_callbacks'</p>
openSUSE admin - tickets #54080 (Resolved): Enable gravatar for vagrant-admins@opensuse.orghttps://progress.opensuse.org/issues/540802019-07-10T12:28:04Zdancermakdcermak@suse.com
<p>Dear openSUSE admins,</p>
<p>I would like to sign up the mailinglist <a href="mailto:vagrant-admins@opensuse.org">vagrant-admins@opensuse.org</a> for<br>
Gravatar, as I need it to display the correct logo on Vagrant Cloud<br>
(this is unfortunately <em>only</em> possible through Gravatar).</p>
<p>I've tried to sign the mailinglist up for a Wordpress account, but the<br>
confirmation email never got through to me. Could you temporarily<br>
disable any Spam filtering or other filters?</p>
<p>Thanks in advance,</p>
<p>Dan</p>
<p>-- <br>
Dan Čermák <a href="mailto:dcermak@suse.com">dcermak@suse.com</a><br>
Software Engineer Development tools<br>
SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany<br>
Tel: +49-911-74053-0; Fax: +49-911-7417755; <a href="https://www.suse.com/" class="external">https://www.suse.com/</a><br>
GF: Felix Imendörffer, Mary Higgins, Sri Rasiah<br>
HRB 21284 (AG Nürnberg)</p>