Jira (PUP-1125) Exec checks should be turned into metaparameters
Title: Message Title Felix Frank commented on PUP-1125 Re: Exec checks should be turned into metaparameters Hi Steve Potter, late reply, I missed your post. Frankly, I don't believe that your use cases necessarily call for these metaparameters. 1) The list of DBvols per server should be part of their Hiera data. Puppet should be your authority for such things. Failing that, this example would warrant a custom fact, I believe. 2) In and of itself, it's tempting to force Puppet to not touch the user once it has been created. Please note the following factors, however: 1. There is a module to manage things only one: https://forge.puppet.com/binford2k/manageonce 2. If you use only_if or manageonce::, all other properties of your user resources will fall out of management as well. This can be undesirable. Granted, the final point is tricky. Off the top of my head, I cannot think of a way to work around it cleanly without falling back to yet another custom fact. Add Comment This message was sent by Atlassian JIRA (v6.4.13#64028-sha1:b7939e9) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-6259) File resource no_proxy behavior is ignored
Title: Message Title
Felix Frank commented on PUP-6259
Re: File resource no_proxy behavior is ignored
Eh, that would probably be gratuitous. I had so far assumed that this was a server certificate issue, but if the proxy returns 403s, that is the reason for your issues of course, and has little to do with other specifics.
It's a long shot, but can you try and run your resource through puppet apply?
puppet apply -e "file {'/var/opt/foo/client.config': ensure => present, source => 'https://files.internal/client.config', }"
Add Comment
This message was sent by Atlassian JIRA (v6.4.13#64028-sha1:b7939e9)
--
You received this message because you are subscribed to the Google Groups "Puppet Bugs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at https://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-6259) File resource no_proxy behavior is ignored
Title: Message Title Felix Frank commented on PUP-6259 Re: File resource no_proxy behavior is ignored How does external HTTPS work in this scenario? I would assume the proxy also interferes with access to, say, forge.puppet.com. Can you puppet module install things on this machine? How are the "global" environment variables managed? Is it certain that the Puppet agent uses them? Add Comment This message was sent by Atlassian JIRA (v6.4.13#64028-sha1:b7939e9) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1113) Ability to display messages during agent run without creating resource changes
Title: Message Title Felix Frank commented on PUP-1113 Re: Ability to display messages during agent run without creating resource changes I suggest this be rejected instead. This module serves the stated purpose: https://forge.puppetlabs.com/ipcrm/echo Personally I don't feel that including this type in the core is warranted. Add Comment This message was sent by Atlassian JIRA (v6.4.12#64027-sha1:e3691cc) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4669) Varargs support for (defined) types (and possibly classes)
Title: Message Title
Felix Frank commented on PUP-4669
Re: Varargs support for (defined) types (and possibly classes)
Consider this example:
create_resources('apache::vhost',hiera('apache_virtual_hosts'))
There are a couple of scenarios that make it desirable to wrap the virtual_host type with a custom define.
define custom::vhost($docroot,*$rest) {
$docroot_real =
apache::vhost { $title: docroot => $docroot_real, * => $rest }
}
This wrapper can be used right in place of apache::vhost:
create_resources('custom::vhost',hiera('apache_virtual_hosts'))
A construct that uses the extra_options pattern, on the other hand, will require the user to change the Hiera data to include an extra_options hash in the parameter list.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank commented on PUP-1072 Re: support HTTP(S) URL as the file 'source' Yes, this is currently a known limitation. For most HTTP(S) sources, the resource will only work properly when using checksum => 'mtime', Add Comment This message was sent by Atlassian JIRA (v6.4.12#64027-sha1:e3691cc) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank assigned an issue to Michael Smith Puppet / PUP-1072 support HTTP(S) URL as the file 'source' Change By: Felix Frank Assignee: Michael Smith Add Comment This message was sent by Atlassian JIRA (v6.4.12#64027-sha1:e3691cc) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank commented on PUP-5395 Re: There's no way to set resolution_type when using data bindings Sounds like a good plan. I think this will even satisfy John Bollinger's request for consistent lookup behavior. Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank commented on PUP-5395 Re: There's no way to set resolution_type when using data bindings My knowledge on YAML is failing me. Does the merge allow the same key to appear multiple times in one hash declaration? I.e. code lookupoptions: mysql_server: ... ... lookupoptions: apache: ... code If this is an issue, then that would be a usability minus compared to the lookupoptions::* keys. Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank commented on PUP-5395 Re: There's no way to set resolution_type when using data bindings True. So yes, I guess we can go with this design instead. Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank updated an issue Puppet / PUP-5395 There's no way to set resolution_type when using data bindings Change By: Felix Frank Labels: hiera parameter redmine Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank updated an issue Puppet / PUP-5395 There's no way to set resolution_type when using data bindings Change By: Felix Frank Component/s: Community Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-5395) There's no way to set resolution_type when using data bindings
Title: Message Title Felix Frank assigned an issue to Rob Nelson Puppet / PUP-5395 There's no way to set resolution_type when using data bindings Change By: Felix Frank Assignee: Andrew Parker Rob Nelson Add Comment This message was sent by Atlassian JIRA (v6.4.11#64026-sha1:78f6ec4) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (HI-184) Inconsistent handling of fully-qualified facts when called from the command line
Title: Message Title Felix Frank assigned an issue to Felix Frank Hiera / HI-184 Inconsistent handling of fully-qualified facts when called from the command line Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4754) Solaris crontabs apparently do not support environment lines
Title: Message Title Felix Frank created an issue Puppet / PUP-4754 Solaris crontabs apparently do not support environment lines Issue Type: Bug Affects Versions: PUP 3.7.0 Assignee: Kylo Ginsberg Components: Types and Providers Created: 2015/06/14 10:41 AM Environment: Solaris Labels: community Priority: Normal Reporter: Felix Frank According to @halfninja on GitHub, cron on Solaris does not support setting environment variables on designated lines. The POSIX standard does not include this capability either. If someone can confirm this, it would perhaps be safest to limit the environment property to agent platforms that support it (i.e., anything but Solaris).
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank commented on PUP-3341 Re: Puppet apply breaks when an ENC returns an environment Most definitely possible. Originally, I wrote that patch for 3.6 I think, but we couldn't agree to merge before master advanced to 4.x. I'll go hunting for the old commits, they might still linger in one repo or another. Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank updated an issue Puppet / PUP-1072 support HTTP(S) URL as the file 'source' Change By: Felix Frank Fix Version/s: PUP4.x Fix Version/s: PUP4.2.0 Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4653) The filebucket client uses v1 API URLs internally
Title: Message Title Felix Frank created an issue Puppet / PUP-4653 The filebucket client uses v1 API URLs internally Issue Type: Improvement Affects Versions: PUP 2.7.23 Assignee: Felix Frank Components: Client Created: 2015/05/26 2:03 PM Fix Versions: PUP 4.2.0 Labels: community indirector Priority: Normal Reporter: Felix Frank Since the 0.25 days, the filebucket client in the form of Puppet::FileBucket::Dipper has interfaces with the indirector by constructing URLs in the following fashion:
Jira (PUP-4619) fstab entries can be written in bad order
Title: Message Title Felix Frank commented on PUP-4619 Re: fstab entries can be written in bad order A variation on this issue is fixed as well: /opt/data02/mysql /var/lib/mysql bind,rw 0 0 /dev/vg0/data2 /opt/data02 ext4 defaults,ro 0 2 The former is a bind mount and must appear after the filesystem on the second line. Puppet would break this order if the latter mount was added later than the bind mount. Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4619) fstab entries can be written in bad order
Title: Message Title Felix Frank created an issue Puppet / PUP-4619 fstab entries can be written in bad order Issue Type: Improvement Affects Versions: PUP 2.7.25 Assignee: Felix Frank Components: Types and Providers Created: 2015/05/17 1:19 PM Environment: *NIX Fix Versions: PUP 4.x Labels: community Priority: Normal Reporter: Felix Frank Given the following fstab content in /tmp/fstab
Jira (PUP-2576) Implement behavioral change in the crontab provider to allow more drastic purging
Title: Message Title Felix Frank commented on PUP-2576 Re: Implement behavioral change in the crontab provider to allow more drastic purging We have no support for cron.d files whatsoever. It was requested in the olden times as https://projects.puppetlabs.com/issues/746 but I don't think this one was migrated, much less implemented. Even if someone would do it now or has already, it would likely exist in a module only. It would be unlikely to be taken into the core. As for purging behavior, I'm actually not sure what happens when multiple providers are operational. But again, this doesn't currently matter because there is no support. Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank commented on PUP-3341 Re: Puppet apply breaks when an ENC returns an environment The --manifest was used for posterity only. In production, it will default to $manifestdir/site.pp or $environmentpath/$environment/manifests/site.pp. The bug causes this default (or whatever :manifest setting is in effect) to be used instead of the manifest that the user is actually trying to apply. Add Comment This message was sent by Atlassian JIRA (v6.3.15#6346-sha1:dbc023d) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4080) support S3 URL as the file type 'source'
Title: Message Title Felix Frank assigned an issue to Eric Sorenson Puppet / PUP-4080 support S3 URL as the file type 'source' Change By: Felix Frank Assignee: EricSorenson Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-4080) support S3 URL as the file type 'source'
Title: Message Title Felix Frank commented on PUP-4080 Re: support S3 URL as the file type 'source' Handing this to Eric Sorenson since this needs a decision. Could perhaps be delegated to Ethan Brown or someone else. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank commented on PUP-1072 Re: support HTTP(S) URL as the file 'source' Meanwhile, mtime is fully functional. Updates server files should get synchronized (smoke testing was promising, although I have yet to go to the length of refreshing an actual webserver). Moreover, if you own the webserver that hosts the file in question, you can enable content digests (apache) so that the md5 sum will be part of the HTTP headers. Puppet will verify this sum as with all files. Since the limitations I described earlier are partly resolved now, adding support for checksums as parameters doesn't seem profitable to me. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3889) Fix puppet agent writing source to file when using non-md5 checksum type
Title: Message Title Felix Frank commented on PUP-3889 Re: Fix puppet agent writing source to file when using non-md5 checksum type Oh I see. Thanks for clearing this up. I guess it's fine - if you know how to read the tickets, it's reasonably simple to locate commits towards the subtasks. After all, there was no one branch to solve the encompassing task. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3889) Fix puppet agent writing source to file when using non-md5 checksum type
Title: Message Title Felix Frank commented on PUP-3889 Re: Fix puppet agent writing source to file when using non-md5 checksum type Was this a duplicate of PUP-1208 ? I'm quite confused, for that one has been marked resolved as well, but I can find no commit that references it. There is also no reference to this ticket, either. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3901) Host type has unexpected behavior and is not idempotent
Title: Message Title Felix Frank commented on PUP-3901 Re: Host type has unexpected behavior and is not idempotent The `clientcert` pseudo-fact takes whatever the client uses for `certname`. This options defaults to the value of the `fqdn` fact, yes. On Linux, I don't think that `hostname` and `fqdn` are strictly related, but they do map do different invocations of `dnsdomainname`, I believe. Controlling them separately is a certain way to madness. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-731) Masters cannot reliably distinguish between multiple versions of a type/function/plugin used in different environments
Title: Message Title Felix Frank updated an issue Puppet / PUP-731 Masters cannot reliably distinguish between multiple versions of a type/function/plugin used in different environments Hi, as this discussion does not pertain to a possible resolution, it would be better to take it to the puppet-users mailing list, actually. But now that we are here, see comments inline. This should work as far as I can tell. The most common failure is updating Ruby code that has been changed during a master process's lifetime, but sure, inconsistencies might also occur. So long as a given process is confined to one environment, it will only suffer stale type/function definitions, no mixups. The whole transaction is restarted, with the agent issuing a new catalog request. After all, the set of available facts might differ among environments. I'm not certain just when custom types are loaded. For custom functions, those are picked up when needed. You could very well end up with different cached functions that are from different points in time, with neither function changing internally until the process terminates. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-731) Masters cannot reliably distinguish between multiple versions of a type/function/plugin used in different environments
Title: Message Title Felix Frank updated an issue Puppet / PUP-731 Masters cannot reliably distinguish between multiple versions of a type/function/plugin used in different environments Hi, as this discussion does not pertain to a possible resolution, it would be better to take it to the puppet-users mailing list, actually. But now that we are here, see comments inline. This should work as far as I can tell. The most common failure is updating Ruby code that has been changed during a master process's lifetime, but sure, inconsistencies might also occur. So long as a given process is confined to one environment, it will only suffer stale type/function definitions, no mixups. The whole transaction is restarted, with the agent issuing a new catalog request. After all, the set of available facts might differ among environments. I'm not certain just when custom types are loaded. For custom functions, those are picked up when needed. You could very well end up with different cached functions that are from different points in time, with neither function changing internally until the process terminates. Add Comment This message was sent by Atlassian JIRA (v6.3.10#6340-sha1:7ea293a) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1295) Yum provider purge target runs irrespective of package installation status
Title: Message Title Felix Frank updated an issue Puppet / PUP-1295 Yum provider purge target runs irrespective of package installation status Change By: Felix Frank Affects Version/s: PUP3.6.1 Affects Version/s: PUP3.2.4 Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1295) Yum provider purge target runs irrespective of package installation status
Title: Message Title Felix Frank updated an issue Puppet / PUP-1295 Yum provider purge target runs irrespective of package installation status Change By: Felix Frank Component/s: TypesandProviders Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1295) Yum provider purge target runs irrespective of package installation status
Title: Message Title
Felix Frank updated an issue
Puppet / PUP-1295
Yum provider purge target runs irrespective of package installation status
Change By:
Felix Frank
Itseemstheyumproviderwilltrytopurgeanabsentpackage: pre {noformat}
Jira (PUP-1208) md5lite, mtime not honoured for file type/provider
Title: Message Title Felix Frank commented on PUP-1208 Re: md5lite, mtime not honoured for file type/provider As discussed briefly during Hackathon: The source support for the file type has some traces of being bolt onto the content management after the fact. There is even some (wanted) inconsistency in behavior among agent and apply, see the discussion on the PUP-1072 PR. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3726) File does not establish sync state when using mtime checksum
Title: Message Title Felix Frank created an issue Puppet / PUP-3726 File does not establish sync state when using mtime checksum Issue Type: Bug Affects Versions: PUP 3.7.3 Assignee: Felix Frank Components: Community, Types and Providers Created: 2014/12/02 5:13 PM Fix Versions: PUP 4.x Priority: Minor Reporter: Felix Frank Original Estimate: 2 days Remaining Estimate: 2 days When creating a file using the
Jira (PUP-1208) md5lite, mtime not honoured for file type/provider
Title: Message Title Felix Frank commented on PUP-1208 Re: md5lite, mtime not honoured for file type/provider I stumbled into this when hacking away at PUP-1072. If the agent is to be able to establish whether a local file is in sync with a file available from a generic HTTP server, it can go by the Last-Modified header vs. the local mtime the Content-Length header vs. the local file size True, there is no way to make sure that the local content is consistent (without of downloading the HTTP resource during each agent transaction, that is). I consider this to be a natural limitation of the concept (HTTP headers don't comprise a content checksum). After receiving file metadata from HTTP headers, the agent should be capable of determining sync if local mtime is equal to or later than server mtime out of sync if local mtime is before server mtime All of that said, HTTP might end up using a custom pseudo-checksum that comprises both mtime and file size for slightly better consistency. As an aside, I'm actually opposed to a 3.7.x fix, since the issue was not introduced in 3.7.0 or later. The fix shoud target master, no? Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) --
Jira (PUP-1043) Specifying file content by checksum should not use 'content' property
Title: Message Title Felix Frank commented on PUP-1043 Re: Specifying file content by checksum should not use 'content' property After thinking some more: Yes, we should aim for `4.0`. It would be great if we could also release another `3.7.x` that will deprecate checksums as `content` values. Users can then bootstrap for the Puppet 4 migration by upgrading to this final `3.7`. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1043) Specifying file content by checksum should not use 'content' property
Title: Message Title Felix Frank updated an issue Puppet / PUP-1043 Specifying file content by checksum should not use 'content' property Change By: Felix Frank Fix Version/s: PUP4.0.0 Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1043) Specifying file content by checksum should not use 'content' property
Title: Message Title Felix Frank updated an issue Puppet / PUP-1043 Specifying file content by checksum should not use 'content' property Change By: Felix Frank Assignee: FelixFrank EricSorenson Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1043) Specifying file content by checksum should not use 'content' property
Title: Message Title Felix Frank commented on PUP-1043 Re: Specifying file content by checksum should not use 'content' property Good point, Erik Daln. I'm not sure that we need to introduce a breaking change. First point of order is the addition of a new file property. Passing checksums as content values will then be deprecated. It can be discontinued at a later date. Then again, seeing as we are breaking cross-version compatibility among master and agent at 4.0 anyway, this would likely be an opportune time to skip the deprecation period and go for the whole change at once, yes. Eric Sorenson, do you think this can make the cut for the 4.0 release? I'm not especially keen on doing it myself - I suspect one of your guys could implement this in a comparatively short time frame. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank updated an issue Puppet / PUP-3341 Puppet apply breaks when an ENC returns an environment Change By: Felix Frank Story Points: 1 2 Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank commented on PUP-1072 Re: support HTTP(S) URL as the file 'source' The sketch of Anthony Ryan is intriguing. I suggest moving it to a separate ticket with the purpose of future improvement. To move this forward, I shall try and look at a KISS way of implementing the support inside the current structure. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1043) Specifying file content by checksum should not use 'content' property
Title: Message Title Felix Frank assigned an issue to Felix Frank Puppet / PUP-1043 Specifying file content by checksum should not use 'content' property Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank commented on PUP-3341 Re: Puppet apply breaks when an ENC returns an environment Josh Cooper mentioned that there seems to be development towards softening the premacy of the ENC wrt. the choice of environments on the master, if the user wants it. I'm not certain how this relates to ENTERPRISE-151. This might mean that this issue will have to be revisited under this shifting premise. I shall go forward with implementing the current behavior. Once the master behavior is adapted, the apply code can follow suit once more. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (FACT-634) Facter emits Already evaluated... warnings
Title: Message Title Felix Frank assigned an issue to Felix Frank Facter / FACT-634 Facter emits Already evaluated... warnings Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (FACT-634) Facter emits Already evaluated... warnings
Title: Message Title Felix Frank commented on FACT-634 Re: Facter emits Already evaluated... warnings Re-opening. I constantly reproduce with facter 2.1, 2.2 and 2.3 on a Debian box with Ruby 1.9.3 (but also saw this with 2.3 and Ruby 2.1). Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (FACT-634) Facter emits Already evaluated... warnings
Title: Message Title Felix Frank commented on FACT-634 Re: Facter emits Already evaluated... warnings Well, what do you know - it is caused by multiple facter versions in Ruby's lookup path. For example, installing the facter package in the operating system will make bundler load facts from both /usr/lib/ruby/vendor_ruby as well as the local source. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3667) The ini_file_spec unit tests need improving
Title: Message Title Felix Frank created an issue Puppet / PUP-3667 The ini_file_spec unit tests need improving Issue Type: Improvement Affects Versions: PUP 3.7.3 Assignee: Felix Frank Components: Community Created: 2014/11/12 1:38 PM Fix Versions: PUP 4.x Labels: tests Priority: Minor Reporter: Felix Frank The unit tests for the inifile support does not seem to use fixtures to their capabilities. Warrants a look at how the tests can be improved through actual I/O and operation.
Jira (PUP-3547) File Server Mounts Should Support Directory Environments
Title: Message Title Felix Frank commented on PUP-3547 Re: File Server Mounts Should Support Directory Environments As far as I know, the ENC will only take effect during the catalog request phase. The master will receive the node data (including environment from the ENC). The ENC is not queried for each and every file retrieval. The environment from the ENC is communicated to the agent as part of the catalog. There is nothing that can stop a rogue agent from ignoring this value. As it stands, you can solve this security issue by carefully maintaining an auth.conf file that has arbitrarily fine grained whitelists that selectively allows agent to access certain fileserver locations. This will be necessary regardless of whether the fileserver supports mounts that are dynamic based on the agent's environment. As such, and I'm sorry to say, this feature (although quite convenient) does not seem to add much value, as far as I can tell. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3547) File Server Mounts Should Support Directory Environments
Title: Message Title Felix Frank commented on PUP-3547 Re: File Server Mounts Should Support Directory Environments From today's PR discussion: Yes, doing this in a secure fashion would require quite a bit of more work in a lot of places. It would also be better suited for adding in puppetserver, because the puppet master based fileserver will be obsolete in the future. Patches for making fileserving more secure would be welcome indeed. As for enhanced capabilities for auth.conf, I suppose that suggestions, designs and discussion (say on puppet-dev) would be most welcome, yes. Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3564) pkg provider latest gives empty warnings
Title: Message Title Felix Frank updated an issue Puppet / PUP-3564 pkg provider latest gives empty warnings Change By: Felix Frank Labels: community Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3564) pkg provider latest gives empty warnings
Title: Message Title Felix Frank updated an issue Puppet / PUP-3564 pkg provider latest gives empty warnings Change By: Felix Frank Component/s: Community Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3564) pkg provider latest gives empty warnings
Title: Message Title Felix Frank updated an issue Puppet / PUP-3564 pkg provider latest gives empty warnings Change By: Felix Frank Assignee: JoshCooper Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3547) File Server Mounts Should Support Directory Environments
Title: Message Title Felix Frank commented on PUP-3547 Re: File Server Mounts Should Support Directory Environments Much clearer, thanks for the write-up! I can't claim to have understood how the combinatorial explosion happens, but I'll just go ahead and trust you on that one. If I understand you correctly, this is mainly supposed to enhance security between environments. It's not really more convenient than using $environment in the fileserver URLs, yes? Are you certain that this is better security? Can the environment in a file request not be forged just in the way that the URL can be manipulated? Add Comment This message was sent by Atlassian JIRA (v6.3.7#6337-sha1:2ed701e) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3547) File Server Mounts Should Support Directory Environments
Title: Message Title Felix Frank commented on an issue Re: File Server Mounts Should Support Directory Environments This doesn't look unreasonable to me, but then it seems that there should be a workaround for this already possible, no? Could r10k not manage some kind of pseudo-module to (sym)link to your actual files? Anything in that vein? Add Comment Puppet / PUP-3547 File Server Mounts Should Support Directory Environments *tl;dr*: I'd like to be able to serve files from the file server and interpolate the agent's environment into the mount path. I use r10k to deploy directory environments for a Puppet Master that manages a small, but growing set of VMware VMs on ESXi. I have a set of files that are supplied by an upstream vendor that do not fit the typical build a modu... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to
Jira (PUP-3547) File Server Mounts Should Support Directory Environments
Title: Message Title Felix Frank updated an issue Puppet / PUP-3547 File Server Mounts Should Support Directory Environments Change By: Felix Frank Assignee: JoshuaM.Keyes Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
Changed the naming algorithm to ignore the key data and just rely on path and a unique counter per file. Each unnamed key gets a unique name.
Gert van Dijk have you had a chance to reproduce any other failures? We're nearing the 3.7.2, which might be the last 3.7 release. If so, further fixes will not be released until Puppet 4.0. Last call
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it,
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
Thanks for the confirm!
Clobbering the whole keys file is not an option. That would get rid of comments and whitespaces, which a user does not expect. (Yay, more regressions.)
Reports for any issues that you find will indeed be extremely helpful. Thanks in advance!
You can group problems that appear similar in one shared ticket, but yes, generally it will help us if issues are separated into distinct tickets (do make them relate to this one).
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
Gert van Dijk can you run from source? Could you try the patch from GH-3164? You'd need to patch your agent, the master is of no consequence.
There is still room for collisions. If you hit that off the spot, then I need to think of something more clever.
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
This will be ugly, but I think I can replicate the workaround that I added to cron back in the day.
generate a unique name for each unnamed line that the provider reads
mark such lines as unnamed on disk
when writing records back to disk, remove the generated names
What's especially unfortunate - the user provider will have to predict the generated names and generate according resources.
Doable but convoluted. I'll make sure to leave enough comments for future generations to be able to dissect the crimes we're committing here.
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
Jira (PUP-1077) Move Nagios types and supporting from core to a module
Title: Message Title Felix Frank commented on an issue Re: Move Nagios types and supporting from core to a module +1 to Eric Sorenson this was exactly my thought. Technically, it might incur a penalty at launch time of each Puppet process, but this ought to be negligible next to the actual work that those processes have to perform (think agent, apply, master). Add Comment Puppet / PUP-1077 Move Nagios types and supporting from core to a module As description says. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title Felix Frank updated an issue Puppet / PUP-3357 Unexpected error with multiple SSH keys without comments Change By: Felix Frank Fix Version/s: 3.7.2 Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3266) Puppet does not recompile catalog after failure
Title: Message Title Felix Frank commented on an issue Re: Puppet does not recompile catalog after failure Thomas Rasmussen this can also occur if you still rely on import statements. Do you? Add Comment Puppet / PUP-3266 Puppet does not recompile catalog after failure I have a setup which works, puppet master compiles catalog and client applies this. Then I make a typo in an included class for my node. Rerun puppet agent on client, it will (sometimes) create an error like this: Error: Could not retrieve catalog from remote server: Error 400 on SERVER: Could not find class Some.class.name for node.name If I t... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank updated an issue Puppet / PUP-3341 Puppet apply breaks when an ENC returns an environment Change By: Felix Frank Fix Version/s: 3.7.2 Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
Yes, using the key is not feasible. It is not uncommon to authorize the same key for different accounts, and Puppet requires unique IDs for all keys.
Charlie Sharpsteen basically has the right of it, but I'm not sure that we can get this to work. Puppet must have a way to target those unmanaged resource for purging, so we would likely need to make Puppet annotate the unnamed keys on disk.
This seems pretty invasive to me, because it might also affect users who don't use purging. (If we can limit this behavior to purging runs, this would be all right, I believe, because all annotated keys will be subject to purging. noop mode would be an unfortunate gray area.)
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
The crux is that the generated resources must be suitable for the ssh_authorized_keys provider to purge the file entries.
We cannot just make up an identification scheme without breaking all manifests out there, because those name the keys by comment.
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email
Jira (PUP-3357) Unexpected error with multiple SSH keys without comments
Title: Message Title
Felix Frank commented on an issue
Re: Unexpected error with multiple SSH keys without comments
Thanks for the input Gert van Dijk. The issue is indeed with the internals of the type system. In a nutshell, system entities (such as lines in authorized_keys files) have to be represented internally in the same fashion that a resource declaration from a manifest does. It must all be consistent.
In theory, we might possibly even get away with replacing the namevar for ssh_authorized_key with a composite value as you suggest (without breaking too many manifests, but still more than we can really get away with), but alas the support for such composites is not quite functional (or so I hear).
Add Comment
Puppet / PUP-3357
Unexpected error with multiple SSH keys without comments
Steps to reproduce: # Enter multple SSH authorized keys for a user without comments (just a key line by line, see example below). # Manage the user with Puppet with the {{purge_ssh_keys}} parameter set to true. # Have Puppet agent 3.7.0+ # Puppet agent run fails completely with an error. (full denial of service of the agent effectively) Expected ...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
Jira (PUP-1755) Inconsistent hash order with Ruby 1.8
Title: Message Title Felix Frank commented on an issue Re: Inconsistent hash order with Ruby 1.8 I still believe that this warrants a ticket against the puppetdb module. If the templates therein expect hash arguments to be ordered, that is unclean implementation. Add Comment Puppet / PUP-1755 Inconsistent hash order with Ruby 1.8 Hash table ordering in Puppet 3.x with Ruby 1.8 is inconsistent. It is known that Ruby 1.8 does not guarantee that hash tables are ordered by insertion order, like 1.9 does. However, whatever ordering Ruby 1.8 uses for a given hash table, it stays consistent, and is always the same whenever you iterate through it. This is not true in Puppet 3.x used wi... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-2846) multiple providers are executed for the same resource
Title: Message Title Felix Frank commented on an issue Re: multiple providers are executed for the same resource I cannot reproduced with 3.6.1. Since I don't have aws installed, I replaced its command with /bin/echo in your provider. This is my output (running as non-root for simplicity). $ puppet apply ./pup-2846.pp Could not retrieve fact='virtual', resolution='anonymous': Permission denied @ rb_sysopen - /sys/firmware/dmi/entries/1-0/raw Notice: Compiled catalog for geras.fritz.box in environment production in 0.21 seconds Error: Execution of '/sbin/iptables -I POSTROUTING 1 -t nat -s 10.1.2.0/24 -o eth0 -p all -m comment --comment 100 snat for network foo2 -j MASQUERADE' returned 3: iptables v1.4.21: can't initialize iptables table `nat': Permission denied (you must be root) Perhaps iptables or your kernel needs to be upgraded. Error: /Stage[main]/Main/Firewall[100 snat for network foo2]/ensure: change from absent to present failed: Execution of '/sbin/iptables -I POSTROUTING 1 -t nat -s 10.1.2.0/24 -o eth0 -p all -m comment --comment 100 snat for network foo2 -j MASQUERADE' returned 3: iptables v1.4.21: can't initialize iptables table `nat': Permission denied (you must be root) Perhaps iptables or your kernel needs to be upgraded. [aws_securitygroup] Exists 000 accept all icmp requests
Jira (PUP-2846) multiple providers are executed for the same resource
Title: Message Title Felix Frank updated an issue Puppet / PUP-2846 multiple providers are executed for the same resource Change By: Felix Frank Assignee: FelixFrank cristifalcas Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-2718) Prefetching is prone to fail for eval_generated resources
Title: Message Title
Felix Frank updated an issue
Puppet / PUP-2718
Prefetching is prone to fail for eval_generated resources
Change By:
Felix Frank
Whenthe{{User}}typeimplementedpurgingof{{ssh_authorized_keys}}usingthe{{eval_generate}}hook,itbecameapparentthatthefollowingmanifest{code}package{'coreutils':before=User[root]} user{'root':purge_ssh_keys='/tmp/keyfile'}{code}wouldnotpurgeanykeysfrom{{/tmp/keyfile}}.Thereasonisthatprefetchingofthekeysfailsinthefollowingfashion:#When{{Transaction#evaluate}}startstraversingthegraph,no{{ssh_authorized_key}}resourcesareyetpartofthegraph#eachresourcehasa{{pre_process}}hookthatinvokes{{#prefetch_if_necessary}}#thishappensforthe{{package}}resourcefirst,and{{prefetch_if_necessary}}invokes{{resources_by_provider}},whichcachesahashofallresourcesinthegraph#duringitsevaluation,the{{user}}resourcegenerates{{ssh_authorized_key}}resourcesforpurging#thesekeysalsoinvoke{{prefetch_if_necesary}},buttheresultfrom{{resources_by_provider}}iscachedandlackstheirreferences#noresourcesarepassedtotheprovider's{{prefetch}}method,theproviderconsiderstheentitiestobeunmanagedandtheresourcesarefalselydeterminedtobealreadyabsentAworkaroundmaycompriseawaytoinvalidatethecacheofthe{{Transaction#resources_by_provider}}methodupon{{eval_generate}}-ingresources.Thisneedsadecision,though.
Add Comment
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at
Jira (PUP-1366) yum provider doesn't respect items provided by a package, only the package name
Title: Message Title
Felix Frank commented on an issue
Re: yum provider doesn't respect items provided by a package, only the package name
Hi Sam Kottler, any update on this?
Add Comment
Puppet / PUP-1366
yum provider doesn't respect items provided by a package, only the package name
A simple use case: In EL5 and earlier, the curl development headers were included in a package called curl-devel. In EL6 and later that package was renamed to libcurl-devel. However, the libcurl-devel package still provides curl-devel. When using something like package { curl-devel: ensure = installed } puppet looks for...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1222) If ensure changes, then no other changes are reported for that resource.
Title: Message Title
Felix Frank commented on an issue
Re: If ensure changes, then no other changes are reported for that resource.
Hi Dan Bode, any update on this?
Add Comment
Puppet / PUP-1222
If ensure changes, then no other changes are reported for that resource.
code: {code} service { 'httpd': ensure = running, enable = true, } {code} even though chkconfig is set to off in both cases, its only logged when the process isn't started. {code} # /sbin/service httpd stop Stopping httpd: [ OK ] [root@dannyboy manifests]# chkconfig httpd off [root@dannybo...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1221) Mixed invocation of parameterized classes leads to order dependencies, should be disallowed
Title: Message Title
Felix Frank commented on an issue
Re: Mixed invocation of parameterized classes leads to order dependencies, should be disallowed
So Henrik Lindberg, I assume this is not on the plate for 4.0, yes?
Add Comment
Puppet / PUP-1221
Mixed invocation of parameterized classes leads to order dependencies, should be disallowed
When a parameterized class has default values for all of its parameters, it may be invoked using either include or class \{ ... \} notation. The include notation is idempotent; the class \{ ... \} notation isn't. As a result, there is an order dependency. This works: {code} class foo($param = defaultvalue) { notify { $param: } } class {...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1221) Mixed invocation of parameterized classes leads to order dependencies, should be disallowed
Title: Message Title Felix Frank updated an issue Puppet / PUP-1221 Mixed invocation of parameterized classes leads to order dependencies, should be disallowed Change By: Felix Frank Affects Version/s: 2.7.23 Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title
Felix Frank updated an issue
Puppet / PUP-1072
support HTTP(S) URL as the file 'source'
Change By:
Felix Frank
Lotsandlotsoffolkswanttobeabletodothis: {code} file{ /tmp/example.txt: source='http://example.com/example.txt' } {code} Thiswouldbegoodtosupport;obviouslythemetadatafortheHTTPURIismuchlessavailablethanviapuppetfileserving,butthiswouldmakealotofpeoplevery,veryhappy.
Add Comment
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1072) support HTTP(S) URL as the file 'source'
Title: Message Title Felix Frank assigned an issue to Felix Frank Puppet / PUP-1072 support HTTP(S) URL as the file 'source' Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3341) Puppet apply breaks when an ENC returns an environment
Title: Message Title Felix Frank created an issue Puppet / PUP-3341 Puppet apply breaks when an ENC returns an environment Issue Type: Bug Affects Versions: 3.7.0 Assignee: Felix Frank Created: 24/Sep/14 4:51 PM Labels: commandline Priority: Normal Reporter: Felix Frank Given the following ENC script (/tmp/enc) #!/bin/sh echo 'classes: []' echo 'environment: production'
Jira (PUP-3258) puppet apply + ENC + 3.7.x: does not read the .pp file
Title: Message Title Felix Frank assigned an issue to Felix Frank Puppet / PUP-3258 puppet apply + ENC + 3.7.x: does not read the .pp file Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3258) puppet apply + ENC + 3.7.x: does not read the .pp file
Title: Message Title
Felix Frank commented on an issue
Re: puppet apply + ENC + 3.7.x: does not read the .pp file
I can reproduce with 3.7.0 and the following.
# /tmp/site.pp
notify { '/tmp/site.pp is in use': }
class foo { notify { 'class foo is included': } }
# ~/.puppet/manifests/site.pp
notify { '~/.puppet/manifests/site.pp is in use': }
#!/bin/sh
# /tmp/enc
cat EOF
classes:
- foo
EOF
This invocation should notify about /tmp/site.pp being in use, but does not.
$ bundle exec puppet apply /tmp/site.pp --node_terminus exec --external_nodes /tmp/enc
...
Notice: ~/.puppet/manifests/site.pp is in use
With 3.6.1, it works as expected.
$ bundle exec puppet apply /tmp/site.pp --node_terminus exec --external_nodes /tmp/enc
...
Notice: /tmp/site.pp is in use
With 3.7.0, it *does* work fine if a plain ENC is used.
$ /tmp/enc /tmp/nodes.yaml
$ bundle exec puppet apply /tmp/site.pp --external_nodes /tmp/nodes.yaml
...
Notice: /tmp/site.pp is in use
Current master 868bf83 is still affected.
Apparently the ENC support broke in some exciting way. I shall bisect between 3.6.1 and 3.7.0, let's see when this happened.
Add Comment
Jira (PUP-3258) puppet apply + ENC + 3.7.x: does not read the .pp file
Title: Message Title Felix Frank commented on an issue Re: puppet apply + ENC + 3.7.x: does not read the .pp file The suspect is commit 53d391b37f0a5f8f5937fcd7584fb6aae6db424b (PUP-2519) Create fallback environment only for default production Looks plausible, digging further. Add Comment Puppet / PUP-3258 puppet apply + ENC + 3.7.x: does not read the .pp file Running 'puppet apply' with an ENC and a .pp file on the command line (full example supplied): at 3.6.1, the 'notice's in the .pp file are executed; at 3.7.x, they are not. 'puppet agent' 3.7.x does the right thing (same as 3.6.1). See the README file in the attached .tar.gz (a complete example). I ran './vapply --noop' (it needed sudo in my case; ma... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To
Jira (PUP-3258) puppet apply + ENC + 3.7.x: does not read the .pp file
Title: Message Title
Felix Frank commented on an issue
Re: puppet apply + ENC + 3.7.x: does not read the .pp file
More to the point, this patch introduces the breakage:
diff --git a/lib/puppet/application/apply.rb b/lib/puppet/application/apply.rb
index 22f2955..bd854e7 100644
--- a/lib/puppet/application/apply.rb
+++ b/lib/puppet/application/apply.rb
@@ -187,7 +187,7 @@ Copyright (c) 2011 Puppet Labs, LLC Licensed under the Apache 2.0 License
configured_environment.override_with(:manifest = manifest) :
configured_environment
-Puppet.override(:environments = Puppet::Environments::Static.new(apply_environment)) do
+Puppet.override(:current_environment = apply_environment) do
# Find our Node
unless node = Puppet::Node.indirection.find(Puppet[:node_name_value])
raise Could not find node #{Puppet[:node_name_value]}
Note that the apply_environment is a simple stub that overrides the manifest setting with what was given on the commandline (apparently). Not wrapping this in a static environment does not seem to be adequate any longer - perhaps the manifest search path must be overridden as well (or something along those lines).
Add Comment
Puppet / PUP-3258
puppet apply + ENC + 3.7.x: does not read the .pp file
Running 'puppet apply' with an ENC and a .pp file on the command line (full example supplied): at 3.6.1, the 'notice's in the .pp file are executed; at 3.7.x, they are not. 'puppet agent' 3.7.x does the right thing (same as 3.6.1). See the README file in the attached .tar.gz (a complete example). I ran './vapply --noop' (it needed sudo in my case; ma...
Jira (PUP-3258) puppet apply + ENC + 3.7.x: does not read the .pp file
Title: Message Title Felix Frank commented on an issue Re: puppet apply + ENC + 3.7.x: does not read the .pp file Hah, bad mojo. This patch seems to fix it: diff --git a/lib/puppet/indirector/node/exec.rb b/lib/puppet/indirector/node/exec.rb index d4faf74..2535ffa 100644 --- a/lib/puppet/indirector/node/exec.rb +++ b/lib/puppet/indirector/node/exec.rb @@ -21,7 +21,7 @@ class Puppet::Node::Exec Puppet::Indirector::Exec # Set the requested environment if it wasn't overridden # If we don't do this it gets set to the local default -result[:environment] ||= request.environment.name +result[:environment] ||= request.environment create_node(request.key, result) end Directory environments don't seem to work with string representations of environments. Looks like Andy Parker fixed this for LDAP back in the day ( PUP-1118 ), so it appears even more plausible that it should be done this way. Add Comment Puppet / PUP-3258 puppet apply + ENC + 3.7.x: does not read the .pp file Running 'puppet apply' with an ENC and a .pp file on the command line (full example supplied): at 3.6.1, the 'notice's in the .pp file are executed; at 3.7.x, they are not. 'puppet agent' 3.7.x does the right thing (same as 3.6.1). See the README file in the attached .tar.gz (a complete example). I ran './vapply --noop' (it needed sudo in my case; ma...
Jira (PUP-1106) Resource refreshes don't check for failed dependencies
Title: Message Title
Felix Frank commented on an issue
Re: Resource refreshes don't check for failed dependencies
That looks quite elegant, Adrien Thebo.
I'm very much tempted to venture back into Redmine and do some reading, because I vaguely remember some endless discussion going on, and I wonder whether we're missing something important now which has been brought up back in the day.
Add Comment
Puppet / PUP-1106
Resource refreshes don't check for failed dependencies
Given this manifest: {code} exec{moo: command = /usr/bin/cowsay 'fail :(', refreshonly = true, logoutput = true, require = Exec[false], subscribe = [ File[/tmp/1], File[/tmp/2], File[/tmp/3] ] } file{/tmp/1: content = 1} file{/tmp/2: content = 2} file{/tmp/3: content = 3} exec{false: command...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to
Jira (PUP-2526) puppet agent retry failed http requests
Title: Message Title Felix Frank commented on an issue Re: puppet agent retry failed http requests In the case of file resources failing due to WAN restrictions, I wonder whether retries are even the best way forward. Wouldn't it be nicer if the agent could be configured to transparently use a nearby blob storage when resolving puppet:/// URLs? Sure, that may have other problems, but it might be worthwhile to at least investigate this line of thought. Marc Seeger, would you care to open a new ticket for the file via WAN issue? Add Comment Puppet / PUP-2526 puppet agent retry failed http requests It would be nice if puppet agent had the ability had the ability to retry failed http requests to the puppet master. I have multiple puppet masters sitting behind a load balancer in AWS (ELB). Whenever we update a puppet master, the node is removed from the load balancer while the update occurs. Unfortunately the AWS ELB does not allow quiescent remova... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the
Jira (PUP-1061) Install multiple package within a single call to the package manager
Title: Message Title Felix Frank commented on an issue Re: Install multiple package within a single call to the package manager Kylo Ginsberg has sketched 7 steps towards this goal. As far as I know, not even one has been implemented yet. I guess the best way forward would be to create a subtask for step 1: Define report schema changes The others should follow, but I don't think it would be wise to create them all on the spot - I agree that we will likely learn some things along the way that will shift our perspective on what needs to happen and how. Needless to say, code contributions would go a long way. Add Comment Puppet / PUP-1061 Install multiple package within a single call to the package manager During the configuration applying process the package manager is called for each package installation. It is possible to reduce the number of calls to the package manager by gathering package installation and delayed some package installation. Naturally, this modification should not break the dependency graph. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
Jira (PUP-3038) Typemanager wastes CPU cycles with global scope defined types
Title: Message Title Felix Frank commented on an issue Re: Typemanager wastes CPU cycles with global scope defined types According to Josh Cooper this may be unnecessary thanks to the feature implemented for PUP-3032. Add Comment Puppet / PUP-3038 Typemanager wastes CPU cycles with global scope defined types Defined type defined in a global scope will make MetaType::Manager go through all known code locations (modules, gems, loadpaths, etc) looking for that_type.rb That combined with a fact that code locations are not cached makes puppet slow when working with large repositories. I'm preparing 2 PRs related to this issue. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-3038) Typemanager wastes CPU cycles with global scope defined types
Title: Message Title Felix Frank commented on an issue Re: Typemanager wastes CPU cycles with global scope defined types Is this related to PUP-1592 ? Add Comment Puppet / PUP-3038 Typemanager wastes CPU cycles with global scope defined types Defined type defined in a global scope will make MetaType::Manager go through all known code locations (modules, gems, loadpaths, etc) looking for that_type.rb That combined with a fact that code locations are not cached makes puppet slow when working with large repositories. I'm preparing 2 PRs related to this issue. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit
Jira (PUP-3038) Typemanager wastes CPU cycles with global scope defined types
Title: Message Title Felix Frank commented on an issue Re: Typemanager wastes CPU cycles with global scope defined types I see, thanks for looking into that. Is defining types outside of modules a frequent occurence? Best practices suggest that all custom classes and defines should be part of modules, with the global scope limited to node blocks and variable assignments. Add Comment Puppet / PUP-3038 Typemanager wastes CPU cycles with global scope defined types Defined type defined in a global scope will make MetaType::Manager go through all known code locations (modules, gems, loadpaths, etc) looking for that_type.rb That combined with a fact that code locations are not cached makes puppet slow when working with large repositories. I'm preparing 2 PRs related to this issue. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to
Jira (PUP-2526) puppet agent retry failed http requests
Title: Message Title Felix Frank commented on an issue Re: puppet agent retry failed http requests Look, I didn't join this discussion to argue semantics or raise this issue on some pedestal of IT Best Practices. Experience mandates certain requirements in the software I use, and I hold at least that standard to the projects I devote my time to. I will not stand for a change in Puppet that may have a negative impact on the user experience. Snide attacks on my argument aside, is there anything disadvantageous about setting this on an opt-in basis? Add Comment Puppet / PUP-2526 puppet agent retry failed http requests It would be nice if puppet agent had the ability had the ability to retry failed http requests to the puppet master. I have multiple puppet masters sitting behind a load balancer in AWS (ELB). Whenever we update a puppet master, the node is removed from the load balancer while the update occurs. Unfortunately the AWS ELB does not allow quiescent remova... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you
Jira (PUP-2526) puppet agent retry failed http requests
Title: Message Title Felix Frank commented on an issue Re: puppet agent retry failed http requests Speaking for myself, I find the feature intriguing, but frankly, I wouldn't want most of my agents to behave that way. The majority of my agents are on the same, reliable network, along with the master. I propose that this is a very common setup. If requests time out, there is probably a rather severe reason for that. I wouldn't want the whole network of agents to potentially contribute to the mayhem by generating more traffic than necessary. So, if Puppet adopts this behavior, I would be most grateful for the ability to turn it off through an agent setting. Can such an option be added? Add Comment Puppet / PUP-2526 puppet agent retry failed http requests It would be nice if puppet agent had the ability had the ability to retry failed http requests to the puppet master. I have multiple puppet masters sitting behind a load balancer in AWS (ELB). Whenever we update a puppet master, the node is removed from the load balancer while the update occurs. Unfortunately the AWS ELB does not allow quiescent remova... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
Jira (PUP-2846) multiple providers are executed for the same resource
Title: Message Title Felix Frank assigned an issue to Felix Frank Puppet / PUP-2846 multiple providers are executed for the same resource Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-2729) reevaluate config_version whenever an agent checks in.
Title: Message Title
Felix Frank updated an issue
Puppet / PUP-2729
reevaluate config_version whenever an agent checks in.
Change By:
Felix Frank
Whenpuppetmanifestsareedited,puppetmasterdusuallynotices,andcomputesanewconfigurationversionbyinvokingtheconfig_versionscriptorbyusingthedefault(derivedfromthecurrenttime).However,whenfilesareedited,theconfigurationversionremainsunchanged. pre {code} #.../modules/mymodule/manifests/init.ppclassmymodule{file{/dir/filename:ensure=file,source=puppet:///mymodule/filename,}} /pre {code} IfIedit {{ .../modules/mymodule/manifests/init.pp }} ,thentheconfigurationversiongetsupdatednexttimeaclientconnects.Sofarsogood.However,ifIedit {{ .../modules/mymodule/files/filename }} ,theconfigurationversionremainsunchanged.Amessageontheclientsay {noformat} info:Applyingconfigurationversion'oldversionnumberhere'. {noformat} Frommypointofview,IwouldlikeittobeseenasasaconfigurationchangewhenIeditafilethatgetsdeployedfromafilessubdirectoryinamodule.Iusesubversionrevisionnumbersintheconfig_version,andIwanttobeabletotellpeoplecheckthattheconfigurationversionisatleastN,tobesurethatyouhavethelatestchangestofilename.
Add Comment
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1060) enablerepo and disablerepo for yum type
Title: Message Title
Felix Frank commented on an issue
Re: enablerepo and disablerepo for yum type
Thanks for taking the time to look into this.
I believe that a regression should be tracked through a new Ticket that refers to this one instead of re-opening.
Add Comment
Puppet / PUP-1060
enablerepo and disablerepo for yum type
it would be nice to be able to enable a disabled repo for the installation on one package. for example installing facter from EPEL. something like; {noformat} package { facter: ensure = installed, enablerepo = [ epel, epel-testing ]; } {noformat}
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at http://groups.google.com/group/puppet-bugs.
For more options, visit https://groups.google.com/d/optout.
Jira (PUP-2576) Implement behavioral change in the crontab provider to allow more drastic purging
Title: Message Title Felix Frank commented on an issue Re: Implement behavioral change in the crontab provider to allow more drastic purging Well, there is a desire to support this style of deprecation in Type code. However, it introduces cost to the manifest compilation for all users, where only few will benefit from it, and for many versions, none will, when no types actually emit any deprecation warnings. That is actually the norm. I'm torn, leaning towards removing the code. Add Comment Puppet / PUP-2576 Implement behavioral change in the crontab provider to allow more drastic purging See PUP-1381 for details on the bug being fixed. This new ticket was created to track the fix itself vs. adding a warning about the change. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at
Jira (PUP-2576) Implement behavioral change in the crontab provider to allow more drastic purging
Title: Message Title Felix Frank commented on an issue Re: Implement behavioral change in the crontab provider to allow more drastic purging So, Josh Cooper or Andy Parker, should a change that has a deprecation warning in 3.7 remove that warning when it's merged into puppet-4? Or will there be a big let's remove all those warnings branch before the 4.0 release? Add Comment Puppet / PUP-2576 Implement behavioral change in the crontab provider to allow more drastic purging See PUP-1381 for details on the bug being fixed. This new ticket was created to track the fix itself vs. adding a warning about the change. This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-1125) Exec checks should be turned into metaparameters
Title: Message Title
Felix Frank commented on an issue
Re: Exec checks should be turned into metaparameters
Puppet 4 sounds reasonable, although it would likely be a non-breaking change, so no sweat. Before the guys chime in, I'd like to respond once more.
John Bollinger, I'm coming around to your view, I guess, but I'd still like to argue some particulars.
Another workaround are Facts, which may not scale to the number of required checks a fact based solution also makes it more difficult to handle dependent resources
Where does that last come from? Surely a fact-based solution makes it easier to deal with dependent resources, because the result of evaluating the condition is embodied in a fact, and can therefore influence multiple resource declarations.
My reasoning was that a Fact bases solution will likely remove resources conditionally
if $::my_custom_fact { package { foo: } }
so that such resources cannot be safely required by others. If I understand you correctly, your point is that the require value can depend on the fact value as well. That is true, albeit unwieldy.
Because right now it is safe to assume that given a
{ 'foo': require = B['bar'] }
A['foo'] will be applied only if all of B['bar']'s properties are in sync. Implementing option 1 would render that assumption unsafe, and that could lead to a world of pain.
I'm torn on this point - if I add an onlyif equivalent to B[bar], I'm setting myself up for this kind of breakage. Or to put it differently, I apparently only care for the resource's properties under certain conditions.
From a design perspective, it would be consistent to make it possible for the user to require such conditional resources and sync the dependent resources even if the condition of the dependency are not met.
Yes, there is room for users to shoot themselves in the foot.
So I do see potential in such a dangerous feature, at least as a design study. Of course, if we add this at any point in time, getting rid of it again would be quite painful (a Puppet 5 kind of retraction).
So a much less dangerous compromise could be to forego the onlyif/unless equivalents for now and only go for a precondition style metaparameter to let resources fail in scriptable circumstances. If I read John's comments correctly, that is what you would prefer as well.
Jira (PUP-1125) Exec checks should be turned into metaparameters
Title: Message Title Felix Frank commented on an issue Re: Exec checks should be turned into metaparameters Let's see if I can summarize the discussion so far. If at all, it would likely be most acceptable to have the metaparameters behave just like the exec parameters That behavior can be approximated by chaining an actual exec before the resource in question note that in contrast to exec, there would be an error message for a resource that is short-circuited in this fashion there are also implications for dependent resources, because the resource in question effectively fails - from this point of view, the workaround is quite dissimilar from the exec parameter's behavior after all Another workaround are Facts, which may not scale to the number of required checks a fact based solution also makes it more difficult to handle dependent resources All things considered, I believe that we did find possible use cases for the metaparameters. Workarounds exist, but are rather clunky. Also, I don't believe that a workaround exists is an adequate reason to deny the implementation of a feature (see nagios config permissions for an example of us walking into that trap before). John's idea for disjunctive defined types is interesting, but I feel that it is hard to grasp - adding complexity where we're currently trying to make things more simple for the user, if possible. To make some progress, I suggest I try and go forward with what I dubbed option 1 in the beginning. Make resources appear as synced under circumstances described by commands passed as metaparameters. For dependency resolution, such resources behave as true constants, seeing as they are in sync for all intents and purposes. I realize that this hardly embodies a consensus decision, but I think it's the least painful way forward and doesn't break anyone's assumptions. If this is not the case after all, please veto below Concerning implementation, I seem to remember that Josh Cooper would see this done at the resource_harness level. Sounds legit to me.
Jira (PUP-2737) user purge_ssh_keys cannot remove keys with spaces in the comment
Title: Message Title Felix Frank assigned an issue to Felix Frank Puppet / PUP-2737 user purge_ssh_keys cannot remove keys with spaces in the comment Change By: Felix Frank Assignee: FelixFrank Add Comment This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups Puppet Bugs group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (PUP-2737) user purge_ssh_keys cannot remove keys with spaces in the comment
Title: Message Title
Felix Frank commented on an issue
Re: user purge_ssh_keys cannot remove keys with spaces in the comment
Oh, so that's valid? Interesting.
It bothered me from the beginning that the purging implements its own trivial parsing logic, and now I'm paying up.
Add Comment
Puppet / PUP-2737
user purge_ssh_keys cannot remove keys with spaces in the comment
Given the following: {noformat} # /root/.ssh/authorized_keys # HEADER: This file was autogenerated at Fri Jun 06 16:59:39 + 2014 # HEADER: by puppet. While it can still be managed manually, it # HEADER: is definitely not recommended. ssh-rsa B3NzaC1yc2EBIwAAAQEA6NF8iallvQVp22WDkTkyrtvp9eWW6A8YVr+kz4TjGYe7gHzIw+niNltGEFHzD8+v1I2YJ6oXev...
This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede)
--
You received this message because you are subscribed to the Google Groups Puppet Bugs group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com.
To post to this group, send email to puppet-bugs@googlegroups.com.
Visit this group at
