Re: [rt-users] "Hidden" tickets suddenly appear
> You need to find out what is preventing the ticket from showing up in > your query. One of your AND conditions may not be met so the ticket > does not show up. Are you saving your Email that is creating the missed > tickets? You could try submitting a similar Email to see if it has the > same visibility problem. Also, do you have any customizations? Sometimes > a change can have unusual ramifications down the line. Another "hidden" ticket appeared last week (11/4) in the search results. It is ~7 months old. * Ticket has zero transactions * Ticket is in 'new' status as expected. Never had a different state. * Ticket is owned by 'Nobody in particular' (aka account name 'Nobody') as expected for a ticket with no transactions. Never had a different owner. * Ticket is in 'atcc-help' queue. Never had a different queue. * All criteria above match the search query below fully: Queue = 'atcc-help' AND ( Status = 'new' OR Status = 'open' OR Status = 'stalled' ) AND ( Owner = 'Nobody' OR Owner = 'jbla...@our.org' ) Yet it sat hidden for ~7 months, not showing in the search results, and then popped into view Friday 11/4. We do not have any homebrew customizations at all for *this* RT instance. We use RT::Extension::Nagios and RT::Extension:Assets. I can't really picture how we could reproduce this failure scenario. -- Jeff Blaine kickflop.net - RT 4.4 and RTIR training sessions, and a new workshop day! https://bestpractical.com/training * Los Angeles - January 9-11 2017
Re: [rt-users] "Hidden" tickets suddenly appear
On 10/26/2016 4:27 AM, Martin Wheldon wrote: > Hi Jeff, > > If you examine the ticket history of the problem ticket it should tell > you what happened > for it to be set back to the "new" status. > > Is it being reopened by a reply to a email from the pervious ticket > corespondance perhaps?? Martin, Right, that's where I first looked. There are zero transactions past the initial creation via email. Nothing has been done to the ticket. Jeff > On 2016-10-25 16:00, Jeff Blaine wrote: >> RT 4.2.13 backed by PostgreSQL, though I feel like we've been >> experiencing the following for many versions. >> >> I've been suspect for quite some time, but thought maybe I was losing my >> mind instead. I've confirmed finally that I am not losing my mind re: RT >> in this specific case. We are definitely seeing the following situation. >> >> I'm curious if anyone has thoughts as to where to start debugging the >> root cause. >> >> I use a simple saved search as my view into RT at work. Others using >> this queue also use a similar one. The query is: >> >> Queue = 'atcc-help' >> AND >>( >> Status = 'new' >> OR Status = 'open' >> OR Status = 'stalled' >>) >> AND ( >>Owner = 'Nobody' >>OR Owner = 'jbla...@our.org' >> ) >> >> The display settings for the search have "Rows: Unlimited" >> >> It results in a daily list of 70-120 tickets. >> >> We are seeing tickets appear in this list, where they have never >> appeared in the list before, after some period of being in the queue. In >> most cases, the mystery ticket has a "Created" date of more than a few >> weeks ago. The most recent case was 5 months. That is, a ticket with >> "Created" of around 5 months ago, and in "new" status, showed up in the >> results of the search above where it never had shown up before. >> >> As you can imagine, this is terrible for customer service. Luckily it >> seems pretty rare. I would estimate that we see this happen ~5 times per >> year, and then have to start the very awkward conversation about how the >> ticket "slipped through our cracks". >> >> Thoughts very welcome. >> >> Jeff >> - >> RT 4.4 and RTIR training sessions, and a new workshop day! >> https://bestpractical.com/training >> * Boston - October 24-26 >> * Los Angeles - Q1 2017 > - > RT 4.4 and RTIR training sessions, and a new workshop day! > https://bestpractical.com/training > * Boston - October 24-26 > * Los Angeles - Q1 2017 > -- Jeff Blaine kickflop.net - RT 4.4 and RTIR training sessions, and a new workshop day! https://bestpractical.com/training * Los Angeles - Q1 2017
[rt-users] "Hidden" tickets suddenly appear
RT 4.2.13 backed by PostgreSQL, though I feel like we've been experiencing the following for many versions. I've been suspect for quite some time, but thought maybe I was losing my mind instead. I've confirmed finally that I am not losing my mind re: RT in this specific case. We are definitely seeing the following situation. I'm curious if anyone has thoughts as to where to start debugging the root cause. I use a simple saved search as my view into RT at work. Others using this queue also use a similar one. The query is: Queue = 'atcc-help' AND ( Status = 'new' OR Status = 'open' OR Status = 'stalled' ) AND ( Owner = 'Nobody' OR Owner = 'jbla...@our.org' ) The display settings for the search have "Rows: Unlimited" It results in a daily list of 70-120 tickets. We are seeing tickets appear in this list, where they have never appeared in the list before, after some period of being in the queue. In most cases, the mystery ticket has a "Created" date of more than a few weeks ago. The most recent case was 5 months. That is, a ticket with "Created" of around 5 months ago, and in "new" status, showed up in the results of the search above where it never had shown up before. As you can imagine, this is terrible for customer service. Luckily it seems pretty rare. I would estimate that we see this happen ~5 times per year, and then have to start the very awkward conversation about how the ticket "slipped through our cracks". Thoughts very welcome. Jeff - RT 4.4 and RTIR training sessions, and a new workshop day! https://bestpractical.com/training * Boston - October 24-26 * Los Angeles - Q1 2017
[rt-users] 4.2.13 rt-crontool search result oddities
I've just started exploring rt-crontool after many years of RT use, and the following results, per the queries, don't make any sense to me. Can someone point out what I am doing wrong? 1. Owner nobody, status new, created more than 2 days ago /rt/bin/rt-crontool --search RT::Search::FromSQL --search-arg "Owner = 'Nobody' AND Status = 'new' AND Created > '2 days ago'" --action RT::Action --verbose Does not match this ticket, but should: Status: new Owner: Nobody in particular Created: Wed Jan 20 12:12:53 2016 2. Owner nobody, status new, created less than 90 days ago /rt/bin/rt-crontool --search RT::Search::FromSQL --search-arg "Owner = 'Nobody' AND Status = 'new' AND Created < '90 days ago'" --action RT::Action --verbose Matches this ticket, but should not: Status: new Owner: Nobody in particular Created: Mon Apr 07 12:18:30 2014 - RT 4.4 and RTIR Training Sessions https://bestpractical.com/training * Los Angeles - September, 2016
Re: [rt-users] Pg FTS query works, RT search returns 0 results ( was: '.' as delimiter/boundary breaks domain name searches)
>> * An RT 4.2.12 search for the same string returns 0 results[2].
>> I'm 98% certain the ticket *is* indexed though as other queries
>> return it[2].
...
> Is the ticket in question a merged ticket? You had a mail from the
> same timeframe which dealt with merged ticket FTS, which is a known
> bug: https://issues.bestpractical.com/Ticket/Display.html?id=9370
> Sadly, there's not good solution to that bug at this time.
I'm losing what little is left of my mind, apparently. I looked last
week at the problem ticket and could swear it had no merging in its
history. Looking today, I see it *does*.
> perl -MRT=-init -le
> '$t=RT::Tickets->new(RT->SystemUser);$t->FromSQL("@ARGV");print
> $t->BuildSelectQuery' \
> "Content LIKE 'f...@domain.com'"
Handy! Used this to generate the SQL. Pasted the SQL into psql for kicks
and got the unfortunately now-expected 0 results. Filed this snippet
away for later.
Thanks Alex.
Jeff
[rt-users] Pg FTS query works, RT search returns 0 results ( was: '.' as delimiter/boundary breaks domain name searches)
I'm reviving this one time in case anyone has further ideas.
* PostgreSQL 8.4.20 (RHEL 6.6) with FTS does the right
thing when parsing an email address[1]
* An RT 4.2.12 search for the same string returns 0 results[2].
I'm 98% certain the ticket *is* indexed though as other queries
return it[2].
* The rt-fulltext-indexer script runs every 10 minutes and has
no errors.
Jeff
Footnote 1:
rt4=# SELECT alias, description, token FROM ts_debug('f...@domain.com');
alias | description | token
---+---+
email | Email address | f...@domain.com
(1 row)
rt4=# SELECT alias, description, token FROM ts_debug('<f...@domain.com>');
alias | description | token
---+---+
blank | Space symbols | <
email | Email address | f...@domain.com
blank | Space symbols | >
(3 rows)
rt4=#
Footnote 2:
Content LIKE 'f...@domain.com'
0 results
Content LIKE 'domain.com'
0 results
Content LIKE 'domain'
0 results
Content LIKE 'another unique string in the ticket I am trying to hit'
1 result
On 10/2/2014 11:08 AM, k...@rice.edu wrote:
> On Thu, Oct 02, 2014 at 10:56:56AM -0400, Kevin Falcone wrote:
>> On Wed, Oct 01, 2014 at 03:50:43PM -0400, Jeff Blaine wrote:
>>> [ Similar, but unrelated to my other message from 10 minutes ago. ]
>>>
>>> It appears any '.' is interpreted as a word boundary with
>>> Pg full-text indexing turned on.
>>>
>>> Is that known to be true, or am I wrong?
>>>
>>> This breaks searches for FQDNs names in ticket contents.
>>>
>>> Searching for 'foobar' will hit foobar.org
>>>
>>> Searching for 'foobar.org' will not hit 'foobar.org'
>>
>> What FTS will match/return is dictated by your database and its
>> configuration.
>>
>> Have you reviewed the Postgres full text search documentation for your
>> release of Pg?
>>
>> http://www.postgresql.org/docs/8.4/static/textsearch.html
>>
>> -kevin
>
>
> Wow! PostgreSQL 8.4, 4 major releases back! I cannot be certain that I
> am recalling this correctly, but the default parser in older versions
> of PostgreSQL did have that behavior. I do not know when they made the
> change to fix it. What do you get when you run:
>
> rt3=# select plainto_tsquery('rice.edu');
> plainto_tsquery
> -
> 'rice.edu'
> (1 row)
>
> I seem to recall that in the older version when I saw this issue, it
> returned:
>
> plainto_tsquery
> -
> 'rice' & 'edu'
>
> You may be able to make a custom config for your text search using
> the definitions from the current release. I just ended up searching
> for 'rice' instead of 'rice.edu', for example.
>
> Regards,
> Ken
>
--
Jeff Blaine
kickflop.net
PGP/GnuPG Key ID: 0x0C8EDD02
[rt-users] Autocomplete vs. Drop-down for 'Owner'
RT 4.2.11 I'm failing to find where in the code this 50 privileged users = use autocompleter logic is. Where can I override this so that the drop-down remains in place? I don't see that 50 token anywhere with a conditional around it. Our value for $AutocompleteOwners is the default, 0, meaning use the drop down until 50 exist We want our 51-60 users to have the drop down always. Someone point me in the right direction in the code, please?
Re: [rt-users] How to make a field mandatory
See http://blog.bestpractical.com/2013/03/new-rt-extension-for-making-fields-mandatory-on-status-transitions.html On 4/27/2015 6:36 AM, ABD EL MALEK BOUBARNOUS wrote: Thank you Sally for your answer, When chosing (?#Mandatory) as a value in the VALIDATION section of the custom field, it actually become mandatory, but the message that appears in the form says something like that : Field must match the mandary field, Is there a possibility to customise that message to be some like 'Mandatory field' Best regards, 2015-04-27 11:29 GMT+01:00 Sally Ainsley sally.ains...@lifecycle-software.com mailto:sally.ains...@lifecycle-software.com: When setting up a Custom Field there is a section called VALIDATION – you can make a field mandatory there. __ __ Regards __ __ Sally Ainsley | Lifecycle Software | __ __ *From:*rt-users [mailto:rt-users-boun...@lists.bestpractical.com mailto:rt-users-boun...@lists.bestpractical.com] *On Behalf Of *ABD EL MALEK BOUBARNOUS *Sent:* 27 April 2015 11:14 *To:* rt-users@lists.bestpractical.com mailto:rt-users@lists.bestpractical.com *Subject:* [rt-users] How to make a field mandatory __ __ Hello everyone, I would like to know how to make a field mandatory in RT, custom field for instance. Thanks in advance, -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02
Re: [rt-users] RT 4.2.10 and ExternalAuth using LDAP
https://metacpan.org/pod/RT::Authen::ExternalAuth::LDAP
says:
filter
The filter to use to match RT users. You must specify
it and it must be a valid LDAP filter encased in parentheses.
For example:
filter = '(objectClass=*)',
On 4/20/2015 5:16 AM, Indrek Paas wrote:
Hi,
I'm setting up an RT server on:
CentOS 7.1 x64
Apache 2.4
PostgreSQL
Perl v5.16.3
Trying to use ExternalAuth to LDAP (Microsoft AD) using these settings
in RT_SiteConfig.pm:
Plugin( RT::Authen::ExternalAuth );
Set( $ExternalAuthPriority, [My_LDAP] );
Set( $ExternalInfoPriority, [My_LDAP] );
Set($ExternalSettings, {
'My_LDAP' = {
'type' = 'ldap',
'server'= '1.1.1.1',
'user' = 'rtbinduser@domain.server',
'pass' = 'rtbinduserpw',
'base' = 'ou=Dom Users,ou=Company
AD,dc=domain,dc=server',
'attr_match_list' = [
'Name',
'EmailAddress',
],
'attr_map' = {
'Name' = 'sAMAccountName',
'EmailAddress' = 'mail',
'Organization' = 'physicalDeliveryOfficeName',
'RealName' = 'cn',
'ExternalAuthId' = 'sAMAccountName',
'Gecos' = 'sAMAccountName',
'WorkPhone' = 'telephoneNumber',
'Address1' = 'streetAddress',
'City' = 'l',
'State' = 'st',
'Zip' = 'postalCode',
'Country' = 'co'
},
},
} );
I start the RT using it's own server : /opt/rt4/sbin/rt-server --port 8080
Page loads in the browser and I can log in as root but when I try to log
in using AD account I see in the logs:
[warning]: Use of uninitialized value $filter in concatenation (.) or
string at
/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm
line 453.
(/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm:451)
I have been digging through google and the LDAP.pm without success. When
I messed with the 'base' value error changed:
[25778] [Mon Apr 20 08:55:33 2015] [warning]: Use of uninitialized value
$filter in concatenation (.) or string at
/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm
line 453.
(/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm:451)
[25778] [Mon Apr 20 08:55:33 2015] [error]: Can't call method
as_string on an undefined value at
/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm
line 357.
Stack:
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth/LDAP.pm:357]
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth.pm:843]
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth.pm:885]
[/opt/rt4/sbin/../lib/RT/User.pm:141]
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/lib/RT/Authen/ExternalAuth.pm:486]
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/html/Elements/DoAuth:10]
[/opt/rt4/local/plugins/RT-Authen-ExternalAuth/html/Callbacks/ExternalAuth/autohandler/Session:1]
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:310]
[/opt/rt4/share/html/autohandler:53]
(/opt/rt4/sbin/../lib/RT/Interface/Web/Handler.pm:208)
Any suggestions to a right direction are welcome.
PS! Went with LDAP because other services on the server use it
successfully. I have installed RT using Kerberos auth before but decided
to use something simpler. :D
--
Indrek
--
Jeff Blaine
kickflop.net
PGP/GnuPG Key ID: 0x0C8EDD02
Re: [rt-users] Improper display of ticket transaction's HTML
An update, below, on this issue in case anyone comes across it. Seems
like a bug to me.
On 3/19/2015 9:16 AM, Jeff Blaine wrote:
On 3/11/2015 3:56 AM, Christian Loos wrote:
Hi,
upgrading to a more recent version of RT and installing HTML::Gumbo
gives you tables in transactions (RT supports this since 4.2.6) [1].
If you can't upgrade, you can use RT::Extension::PermissiveHTMLMail [2].
Chris
[1] https://www.bestpractical.com/docs/rt/4.2/RT_Config.html#PreferRichText
[2] https://github.com/bestpractical/rt-extension-permissivehtmlmail
GOOD:
We've updated our test server to RT 4.2.10 and installed HTML::Gumbo per
the notes in the documentation for $PreferRichText. When
local/lib/RT/Interface/Web_Local.pm is configured to allow various table
attributes, including 'style', the transaction HTML now displays as a
table properly!
BAD:
But now there are various html, head, body tags on what seems to be only
search result blocks.
IIRC, image attachments are not allowed to the list. Here are 2
screenshots showing what I am talking about in case anyone wants to see:
https://www.dropbox.com/s/euca82h6qfagki1/rt-4210-broken.jpg?dl=0
https://www.dropbox.com/s/ym2ctowlo8qhmnf/rt-4210-broken2.jpg?dl=0
This problem goes away if I uninstall HTML::Gumbo
Am 10.03.2015 um 20:35 schrieb Jeff Blaine:
Hi,
We're using RT 4.2.5 (for now).
$PreferRichText is turned on.
We've put in place a Web_Local.pm that is allowing all tags via:
push @SCRUBBER_ALLOWED_TAGS, qw('*');
Something is still mucking with our HTML though, specifically (that
we've found so far) various table tags' attributes like borders and
padding seem to not be rendered properly. The table borders don't render
at all.
Any ideas?
It is a little unclear to me exactly how to use
%SCRUBBER_ALLOWED_ATTRIBS ... setting it to {'*', 1} caused various
links in the web UI to not work anymore.
--
Jeff Blaine
kickflop.net
PGP/GnuPG Key ID: 0x0C8EDD02
Re: [rt-users] Improper display of ticket transaction's HTML
On 3/11/2015 3:56 AM, Christian Loos wrote:
Hi,
upgrading to a more recent version of RT and installing HTML::Gumbo
gives you tables in transactions (RT supports this since 4.2.6) [1].
If you can't upgrade, you can use RT::Extension::PermissiveHTMLMail [2].
Chris
[1] https://www.bestpractical.com/docs/rt/4.2/RT_Config.html#PreferRichText
[2] https://github.com/bestpractical/rt-extension-permissivehtmlmail
GOOD:
We've updated our test server to RT 4.2.10 and installed HTML::Gumbo per
the notes in the documentation for $PreferRichText. When
local/lib/RT/Interface/Web_Local.pm is configured to allow various table
attributes, including 'style', the transaction HTML now displays as a
table properly!
BAD:
But now there are various html, head, body tags on what seems to be only
search result blocks.
IIRC, image attachments are not allowed to the list. Here are 2
screenshots showing what I am talking about in case anyone wants to see:
https://www.dropbox.com/s/euca82h6qfagki1/rt-4210-broken.jpg?dl=0
https://www.dropbox.com/s/ym2ctowlo8qhmnf/rt-4210-broken2.jpg?dl=0
Am 10.03.2015 um 20:35 schrieb Jeff Blaine:
Hi,
We're using RT 4.2.5 (for now).
$PreferRichText is turned on.
We've put in place a Web_Local.pm that is allowing all tags via:
push @SCRUBBER_ALLOWED_TAGS, qw('*');
Something is still mucking with our HTML though, specifically (that
we've found so far) various table tags' attributes like borders and
padding seem to not be rendered properly. The table borders don't render
at all.
Any ideas?
It is a little unclear to me exactly how to use
%SCRUBBER_ALLOWED_ATTRIBS ... setting it to {'*', 1} caused various
links in the web UI to not work anymore.
--
Jeff Blaine
kickflop.net
PGP/GnuPG Key ID: 0x0C8EDD02
[rt-users] Improper display of ticket transaction's HTML
Hi,
We're using RT 4.2.5 (for now).
$PreferRichText is turned on.
We've put in place a Web_Local.pm that is allowing all tags via:
push @SCRUBBER_ALLOWED_TAGS, qw('*');
Something is still mucking with our HTML though, specifically (that
we've found so far) various table tags' attributes like borders and
padding seem to not be rendered properly. The table borders don't render
at all.
Any ideas?
It is a little unclear to me exactly how to use
%SCRUBBER_ALLOWED_ATTRIBS ... setting it to {'*', 1} caused various
links in the web UI to not work anymore.
[rt-users] *Some* attachments not clickable links?
We're experiencing something odd. Some tickets are not allowing the attachments under Attachments to be clicked in any way. Just the base filename is displayed. Some (most) tickets work fine. Any ideas for which rabbit hole to head down in order to figure this out? RT 4.2.5
Re: [rt-users] Supress owner change emails?
On 11/6/2014 5:43 AM, Johan Sundström wrote: Hi! We have set up a script that changes the owner of a ticket upon first email-reply to the ticket. When the owner is changed, RT sends out a mail about this(see below). We would like to disable this notice from beeing sent, any ides on how to do this? Admin - Scrips - Select - On Owner Change Notify Owner Uncheck enabled, click Save Changes -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02
[rt-users] Pg, full-text indexing, getting search misses
RT 4.2.5 PostgreSQL 8.4 from RHEL 6.x In production, working fine in general 1. Ticket 100 contained (as text/plain) the string 'f...@bar.com' 2. Ticket 100 was merged into ticket 999 3. rt-fulltext-indexer *has run* (it runs every 10 minutes) 4. rt-fulltext-indexer --all --debug has no output PROBLEM: This web UI search does not return ticket 999 [Content] [matches][f...@bar.com] Any thoughts? Bug to report? -- RT Training November 4 5 Los Angeles http://bestpractical.com/training
[rt-users] '.' as delimiter/boundary breaks domain name searches
[ Similar, but unrelated to my other message from 10 minutes ago. ] It appears any '.' is interpreted as a word boundary with Pg full-text indexing turned on. Is that known to be true, or am I wrong? This breaks searches for FQDNs names in ticket contents. Searching for 'foobar' will hit foobar.org Searching for 'foobar.org' will not hit 'foobar.org' -- RT Training November 4 5 Los Angeles http://bestpractical.com/training
[rt-users] Gravatar-inclusive theme or similar?
Connecting with humans is important. Faces mean something. I looked around and could not find anything - has anyone developed a theme (probably a plugin would be required I guess) that includes Gravatars or similar in at least ticket display? -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] how do you write to syslog?
On 8/22/2014 12:03 PM, Al Joslin wrote: I'm in Custom Scrip and I can't get anything to show syslog I've this line in my scrip syslog(critical, Scrip: On Create Copy Custom Fields From Parent); and nothing shows in the syslog I dn't get it (so to speak) -- what am I doing wrong ? See http://requesttracker.wikia.com/wiki/GlobalObjects -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] SOLVED: Can't locate object method host
On 8/13/2014 6:24 AM, Alex Peters wrote: It looks to me like your URI Perl module might not be properly installed, or might be extremely old. What version does this command report?— $ perl -e1 -M'URI 999' RT's make testdeps should have ensured that the minimum required version of this module (1.59) was already installed. Even though I had a RT-legit version 1.61, 'cpan -i URI' solved the issue (which gave me 1.64). Hmm. -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] Can't locate object method host
On 8/13/2014 8:33 PM, Alex Peters wrote:
What's your WebURL config setting as reported by RT's System
Configuration page?
It looks like that might be malformed.
WebBaseURL 'https://alms2.foo.com'
core config
WebDefaultStylesheet'rudder'
core config
WebDomain 'alms2.foo.com'
site config
WebFlushDbCacheEveryRequest 1
core config
WebHttpOnlyCookies 1
core config
WebImagesURL'/static/images/'
core config
WebNoAuthRegex qr/^ (?:\/+NoAuth\/ | \/+REST\/\d+\.\d+\/NoAuth\/) /x
core config
WebPath ''
core config
WebPort '443'
site config
WebRemoteUserContinuous 1
core config
WebSecureCookies0
core config
WebSessionProperties{}
core config
WebURL 'https://alms2.foo.com/'
core config
On 14/08/2014 10:06 am, Jeff Blaine jbla...@kickflop.net
mailto:jbla...@kickflop.net wrote:
On 8/13/2014 9:24 AM, Kevin Falcone wrote:
On Tue, Aug 12, 2014 at 12:32:18PM -0400, Jeff Blaine wrote:
I keep getting the following errors when using the web UI. Here's one
case just now where I logged in as root at https://alms2.foo.com/ and
was then immediately thrown an error:
[15332] [Tue Aug 12 15:55:24 2014] [info]: Successful login
for root
from 1xx.xx.x.165 (/opt/rt4/sbin/../lib/RT/Interface/Web.pm:816)
[15332] [Tue Aug 12 15:55:24 2014] [error]: Can't locate object
method host via package URI::_foreign at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line 935.
Visiting https://alms2.mitre.org/ shows I *did* login as root though.
Previously, Kevin Falcone had said:
This is RT trying to issue the redirect after login, what
URL did
you use to access RT and what was the next parameter.
I had replied:
I'm not sure what you mean by what was the next parameter.
Look at the code that calls Redirect (which is where your error comes
from).
https://github.com/bestpractical/rt-authen-externalauth/blob/master/html/Elements/DoAuth#L19
What's $next for your user that errors out.
You should also always turn on stacktraces when debugging something
like this.
-kevin
Thanks for the replies, Kevin and Alex P.
The installed URI version is 1.61.
Here's a login and the error it causes with stacktraces on.
Aug 13 19:58:31 alms2 RT: [27927] Successful login for root from
xxx.xx.0.231#012Trace begun at /opt/rt4/sbin
/../lib/RT.pm line
298#012Log::Dispatch::__ANON__('Log::Dispatch=HASH(0x1e0e6c8)',
'Successful login for root from xxx.xx.0.231') called at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line
816#012RT::Interface::Web::AttemptPasswordAuthentication('HASH(0x7485080)')
called at /opt/rt4/share/html/NoAuth/Login.html line
49#012HTML::Mason::Commands::__ANON__('pass', 'REDACTEDPASSWORD',
'next', '7809d15aff10a9764adc454707dbf271', 'user', 'root') called at
/usr/local/share/perl5/HTML/Mason/Component.pm line
138#012HTML::Mason::Component::run('HTML::Mason::Component::FileBased=HASH(0x759be50)',
'pass', 'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271',
'user', 'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1305#012eval {...} at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1295#012HTML::Mason::Request::comp(undef, undef, undef, 'pass',
'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271', 'user',
'root') called at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
600#012RT::Interface::Web::MaybeShowNoAuthPage('HASH(0x75fa250)') called
at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
315#012RT::Interface::Web::HandleRequest('HASH(0x75fa250)') called at
/opt/rt4/share/html/autohandler line
53#012HTML::Mason::Commands::__ANON__('pass', 'REDACTEDPASSWORD',
'next', '7809d15aff10a9764adc454707dbf271', 'user', 'root') called at
/usr/local/share/perl5/HTML/Mason/Component.pm line
138#012HTML::Mason::Component::run('HTML::Mason::Component::FileBased=HASH(0x75d6f68)',
'pass', 'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271',
'user', 'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1300#012eval {...} at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1295#012HTML::Mason::Request::comp(undef, undef, undef, 'pass',
'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271', 'user',
'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm line
484#012eval {...} at /usr/local/share/perl5/HT
Aug 13 19:58:31 alms2 RT: [27927] Can't locate object method host via
package URI::_foreign at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
935.#012#012Stack:#012
[/opt/rt4/sbin
Re: [rt-users] Can't locate object method host
On 8/13/2014 9:24 AM, Kevin Falcone wrote:
On Tue, Aug 12, 2014 at 12:32:18PM -0400, Jeff Blaine wrote:
I keep getting the following errors when using the web UI. Here's one
case just now where I logged in as root at https://alms2.foo.com/ and
was then immediately thrown an error:
[15332] [Tue Aug 12 15:55:24 2014] [info]: Successful login for root
from 1xx.xx.x.165 (/opt/rt4/sbin/../lib/RT/Interface/Web.pm:816)
[15332] [Tue Aug 12 15:55:24 2014] [error]: Can't locate object
method host via package URI::_foreign at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line 935.
Visiting https://alms2.mitre.org/ shows I *did* login as root though.
Previously, Kevin Falcone had said:
This is RT trying to issue the redirect after login, what URL did
you use to access RT and what was the next parameter.
I had replied:
I'm not sure what you mean by what was the next parameter.
Look at the code that calls Redirect (which is where your error comes
from).
https://github.com/bestpractical/rt-authen-externalauth/blob/master/html/Elements/DoAuth#L19
What's $next for your user that errors out.
You should also always turn on stacktraces when debugging something
like this.
-kevin
Thanks for the replies, Kevin and Alex P.
The installed URI version is 1.61.
Here's a login and the error it causes with stacktraces on.
Aug 13 19:58:31 alms2 RT: [27927] Successful login for root from
xxx.xx.0.231#012Trace begun at /opt/rt4/sbin
/../lib/RT.pm line
298#012Log::Dispatch::__ANON__('Log::Dispatch=HASH(0x1e0e6c8)',
'Successful login for root from xxx.xx.0.231') called at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line
816#012RT::Interface::Web::AttemptPasswordAuthentication('HASH(0x7485080)')
called at /opt/rt4/share/html/NoAuth/Login.html line
49#012HTML::Mason::Commands::__ANON__('pass', 'REDACTEDPASSWORD',
'next', '7809d15aff10a9764adc454707dbf271', 'user', 'root') called at
/usr/local/share/perl5/HTML/Mason/Component.pm line
138#012HTML::Mason::Component::run('HTML::Mason::Component::FileBased=HASH(0x759be50)',
'pass', 'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271',
'user', 'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1305#012eval {...} at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1295#012HTML::Mason::Request::comp(undef, undef, undef, 'pass',
'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271', 'user',
'root') called at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
600#012RT::Interface::Web::MaybeShowNoAuthPage('HASH(0x75fa250)') called
at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
315#012RT::Interface::Web::HandleRequest('HASH(0x75fa250)') called at
/opt/rt4/share/html/autohandler line
53#012HTML::Mason::Commands::__ANON__('pass', 'REDACTEDPASSWORD',
'next', '7809d15aff10a9764adc454707dbf271', 'user', 'root') called at
/usr/local/share/perl5/HTML/Mason/Component.pm line
138#012HTML::Mason::Component::run('HTML::Mason::Component::FileBased=HASH(0x75d6f68)',
'pass', 'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271',
'user', 'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1300#012eval {...} at /usr/local/share/perl5/HTML/Mason/Request.pm
line 1295#012HTML::Mason::Request::comp(undef, undef, undef, 'pass',
'REDACTEDPASSWORD', 'next', '7809d15aff10a9764adc454707dbf271', 'user',
'root') called at /usr/local/share/perl5/HTML/Mason/Request.pm line
484#012eval {...} at /usr/local/share/perl5/HT
Aug 13 19:58:31 alms2 RT: [27927] Can't locate object method host via
package URI::_foreign at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line
935.#012#012Stack:#012
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:935]#012
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:830]#012
[/opt/rt4/share/html/NoAuth/Login.html:49]#012
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:600]#012
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:315]#012
[/opt/rt4/share/html/autohandler:53]#012Trace begun at
/opt/rt4/sbin/../lib/RT.pm line
298#012Log::Dispatch::__ANON__('Log::Dispatch=HASH(0x1e0e6c8)', 'Can\'t
locate object method host via package URI::_foreign at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line 935.^J^JStack:^J
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:935]^J
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:830]^J
[/opt/rt4/share/html/NoAuth/Login.html:49]^J
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:600]^J
[/opt/rt4/sbin/../lib/RT/Interface/Web.pm:315]^J
[/opt/rt4/share/html/autohandler:53]^J') called at
/opt/rt4/sbin/../lib/RT/Interface/Web/Handler.pm line
209#012HTML::Mason::Exception::as_rt_error('HTML::Mason::Exception=HASH(0x78f6ee8)')
called at /usr/local/share/perl5/HTML/Mason/Exceptions.pm line
282#012HTML::Mason::Exception::as_string('HTML::Mason::Exception=HASH(0x78f6ee8)',
undef, '') called at /usr/local/share/perl5/HTML/Mason/Request.pm line
570#012HTML::Mason::Request::_handle_error('RT::Interface::Web::Request=HASH(0x73a54b8)',
'HTML::Mason::Exception=HASH(0x78f6ee8)') called at
/usr/local/share/perl5/HTML/Mason/Request.pm
[rt-users] Can't locate object method host
I mentioned this as part of a separate post as I thought it was related
to my other issues in that post (SSL), but it is not.
I keep getting the following errors when using the web UI. Here's one
case just now where I logged in as root at https://alms2.foo.com/ and
was then immediately thrown an error:
[15332] [Tue Aug 12 15:55:24 2014] [info]: Successful login for root
from 1xx.xx.x.165 (/opt/rt4/sbin/../lib/RT/Interface/Web.pm:816)
[15332] [Tue Aug 12 15:55:24 2014] [error]: Can't locate object
method host via package URI::_foreign at
/opt/rt4/sbin/../lib/RT/Interface/Web.pm line 935.
Visiting https://alms2.mitre.org/ shows I *did* login as root though.
Previously, Kevin Falcone had said:
This is RT trying to issue the redirect after login, what URL did
you use to access RT and what was the next parameter.
I had replied:
I'm not sure what you mean by what was the next parameter.
The site was accessed (then and just now) directly as
https://alms2.foo.com/
The browser just reports the generic An internal RT error
occurred. Your administrator... (etc).
And the thread went the direction of the real other problem (SSL).
Thoughts?
# Versions
Apache httpd 2.2.15-31 (RHEL 6.5 package)
RT 4.2.6
RT-Authen-ExternalAuth 0.21
mod_fastcgi 2.4.6
# The only relevant thing I can see in my RT_SiteConfig.pm
Set($WebDomain, 'alms2.foo.com');
Set($WebPort, '443');
# Apache config
LoadModule fastcgi_module modules/mod_fastcgi.so
FastCgiIpcDir /tmp
VirtualHost alms2.foo.com:80
# Redirect 80 to 443
ServerName alms2.foo.com
RewriteEngine On
RewriteOptions Inherit
RewriteCond %{HTTPS} off
RewriteRule (.*) https://%{HTTP_HOST}%{REQUEST_URI}
/VirtualHost
VirtualHost alms2.foo.com:443
ServerName alms2.foo.com:443
#
# Lots of SSL configuration redacted here.
#
FastCgiServer /opt/rt4/sbin/rt-server.fcgi -processes 5 -idle-timeout 300
DocumentRoot /opt/rt4/share/html
Alias /NoAuth/images/ /opt/rt4/share/html/NoAuth/images
Location /NoAuth/images
SetHandler default-handler
/Location
ScriptAlias / /opt/rt4/sbin/rt-server.fcgi/
AddDefaultCharset UTF-8
Location /
Order allow,deny
Allow from all
Options +ExecCGI
AddHandler fcgid-script fcgi
/Location
/VirtualHost
--
RT Training - Boston, September 9-10
http://bestpractical.com/training
Re: [rt-users] 'content matches' and 'content doesn't match' give same results
Can you explain your use case a bit? I expect that you meant it as exclude tickets which do contain? That's correct. -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] 'content matches' and 'content doesn't match' give same results
[ I ran a search with StatementLogging enabled and this ] [ is the sql statement with content not like 'foo.com' ] [ and content like 'foo.com', they are the same. ] [ --Thanks for that, Joop! ] Alex, to be clear, these queries below were done via the search form. I am only quoting the SearchBuilder terms for simplicity. The form stated: [ Content ] [ matches ] __foo.com___ And [ Content ] [ doesn't match ] __foo.com___ and I clicked search to get the same 3 tickets as results for both. I can't see how, in some way or another, this is not a bug. On 8/8/2014 8:38 PM, Alex Peters wrote: Values need to be quoted, I believe. Compare these two queries: Content LIKE foo.com http://foo.com Content LIKE 'foo.com http://foo.com' I wonder whether wildcards might be necessary, like they seem to be in regular SQL: Content LIKE '%foo.com http://foo.com%' On 09/08/2014 12:36 am, Jeff Blaine jbla...@kickflop.net mailto:jbla...@kickflop.net wrote: Setup: RT 4.2.5 with PostgreSQL and full-text indexing enabled and completed for all tickets. Tickets 1, 2, and 3 all have contents with the string foo.com http://foo.com Searching for 'content LIKE foo.com http://foo.com' returns tickets 1, 2, 3 Searching for 'content NOT LIKE foo.com http://foo.com' returns tickets 1, 2, 3 Has anyone seen this? Any ideas what might be going wrong or how do start debugging this? -- Jeff Blaine kickflop.net http://kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] 'content matches' and 'content doesn't match' give same results
Setup: RT 4.2.5 with PostgreSQL and full-text indexing enabled and completed for all tickets. Tickets 1, 2, and 3 all have contents with the string foo.com Searching for 'content LIKE foo.com' returns tickets 1, 2, 3 Searching for 'content NOT LIKE foo.com' returns tickets 1, 2, 3 Has anyone seen this? Any ideas what might be going wrong or how do start debugging this? -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] Some users being forced to reauthenticate more often than others
RT 4.2.5 We're trying to track down an issue where some users are being forced to reauthenticate to RT 3-4 times per day while others aren't. This RT instance is using the builtin authentication targeting the 'rt4' database. Our session-cleaning cron job is as follows 0 * * * * /apps/rt4/sbin/rt-clean-sessions --older 2D Any suggested paths to start down would be welcome. What might cause this? -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] Coredumps, traceback, etc : 4.2.6 and RT-Authen-ExternalAuth
On 8/4/2014 10:50 AM, Kevin Falcone wrote: On Fri, Aug 01, 2014 at 03:14:52PM -0400, Jeff Blaine wrote: RHEL 6.5 RT 4.2.6 RT-Authen-ExternalAuth 0.21 with SSL + TLS on. Using RHEL 6.5 packages, one fails the Net-SSLeay strict version requirement required by RT-Authen-ExternalAuth. RHEL 6.5 comes with a 9th iteration of Net-SSLeay 1.35 (the package name is perl-Net-SSLeay-1.35-9). I don't understand what strict means. I just mean that RT-Authen-ExternalAuth is requiring a specific CPAN version of Net-SSLeay and that RHEL perl-Net-SSLeay-1.35-9 may include that same actually required functionality (given the backporting RH often does for security fixes, etc). I am not suggesting with that explanation that RT or the extension should care or try to cater to various distros' packaging+patching workflows. I'm just explaining what I meant by strict. Using CPAN to install a more modern one, 1.65, results in httpd dumping core at restart time. Have a look at other reports of this, which we've never been able to replicate. Alex replied to one just last week, especially relevant if you're running mod_perl (which you don't say). Yes, we're using RHEL 6.5's Apache httpd and RHEL 6.5's mod_perl. I'll visit the archives. *Disabling* all SSL and TLS for $ExternalSettings in RT_SiteConfig.pm results in the following missing host method error after apparent success: Aug 1 15:01:39 alms2 RT: [24103] Can't locate object method host via package URI::_foreign at /opt/rt4/sbin/../lib/RT/Interface/Web.pm line 935.#012#012Stack:#012 [/opt/rt4/sbin/../lib/RT/Interface/Web.pm:935]#012 [/opt/rt4/local/plugins/RT-Authen-ExternalAuth/html/Elements/DoAuth:19]#012 [/opt/rt4/local/plugins/RT-Authen-ExternalAuth/html/Callbacks/ExternalAuth/autohandler/Session:1]#012 [/opt/rt4/sbin/../lib/RT/Interface/Web.pm:311]#012 [/opt/rt4/share/html/autohandler:53] Any thoughts would very welcome at this point. This is RT trying to issue the redirect after login, what URL did you use to access RT and what was the next parameter. I'm not sure what you mean by what was the next parameter. The site was accessed (then and just now) directly as https://alms2.foo.com/ The browser just reports the generic An internal RT error occurred. Your administrator... (etc). -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] Coredumps, traceback, etc : 4.2.6 and RT-Authen-ExternalAuth
Using CPAN to install a more modern one, 1.65, results in httpd dumping core at restart time. Have a look at other reports of this, which we've never been able to replicate. Alex replied to one just last week, especially relevant if you're running mod_perl (which you don't say). Retaining use of mod_perl, but disabling mod_ssl and leaving the RT-Authen-ExternalAuth set to use SSL/TLS worked. I'm not sure this is a *mod_perl* issue, for Alex's and/or Best Practical's records. mod_ssl + mod_perl + RT-Authen-ExternalAuth with SSL all worked in 2012 when I tested it last :( -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] Delete dashboard? Delete saved search?
I see no web UI method to delete either a dashboard or a saved search. Am I missing it? -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] Told in searches
Shouldn't Told '48 hours ago' return the tickets that have NEVER even been Told? It doesn't as far as I can see. I'm failing to find a way to search for tickets that have never been told. RT 4.2.3. Thoughts? -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] Some users can't see comment / editor box?
For the next person, this ended up being stale RT sessions. Running the RT session cleaning script solved the issue. It had never been set up as a cron job. Thanks for the replies, as always. On 7/10/2014 8:41 PM, Alex Peters wrote: If only some users are affected then it's almost definitely a browser issue, as RT doesn't alter its output per browser to my knowledge (unless RT thinks it's a mobile browser). Clearing browser caches may help. On 11/07/2014 3:23 am, Jeff Blaine jbla...@kickflop.net mailto:jbla...@kickflop.net wrote: We're experiencing something odd with 4.2.x (upgrade from 3.8.17) where SOME users cannot see the text editor box upon, for example, clicking Comment. This is occuring, for *those affected users*, with Firefox on OSX, Firefox + Chrome + IE on Windows. For other users, using what I understand to be the same browsers, it works fine. The affected users did not have any issue commenting with RT 3.8.17 and are in a group that has the CommentOnTicket right for the queue in question. In fact, that group has all General rights for the queue. I would greatly welcome any thoughts before I leave for vacation in 11 hours if anyone has them :) -- Jeff Blaine kickflop.net http://kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] Where are user prefs stored?
Where are user preferences stored? I can't find them in the database. RT 4.2.5 -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
[rt-users] Some users can't see comment / editor box?
We're experiencing something odd with 4.2.x (upgrade from 3.8.17) where SOME users cannot see the text editor box upon, for example, clicking Comment. This is occuring, for *those affected users*, with Firefox on OSX, Firefox + Chrome + IE on Windows. For other users, using what I understand to be the same browsers, it works fine. The affected users did not have any issue commenting with RT 3.8.17 and are in a group that has the CommentOnTicket right for the queue in question. In fact, that group has all General rights for the queue. I would greatly welcome any thoughts before I leave for vacation in 11 hours if anyone has them :) -- Jeff Blaine kickflop.net PGP/GnuPG Key ID: 0x0C8EDD02 -- RT Training - Boston, September 9-10 http://bestpractical.com/training
Re: [rt-users] Missed ticket hits in search
Something curious about the tickets that do not show up in the search results message list. They both show: X-MIME-Autoconverted: from quoted-printable to 8bit in the original message headers. The ticket ACLs are fine (Thanks Ken) On 4/15/2014 2:24 PM, k...@rice.edu wrote: On Tue, Apr 15, 2014 at 02:17:45PM -0400, Jeff Blaine wrote: [ Yes, I know RT 3.8 is EOL :) ] One of our RT instances is still running RT 3.8.17. A colleague reported, and I confirmed carefully with my own eyes, that a search is missing tickets it should be finding. Content LIKE '34.100.108.26' AND Created '2014-04-01' AND Queue = 'HJLAM' This returns 2 results. There are at least 2 other tickets that match all 3 items in the search criteria and are not part of the RT search results. Has anyone seen this before? Are there known oddities? The RT instance is using MySQL. Any suggestions? Thanks, Jeff Hi Jeff, Have you checked the ACLs for all of the tickets? If the user cannot see a ticket, then it will not be returned in a search. It does not matter that the search criteria match if the account does not have the rights to see the ticket. Regards, Ken -- RT Training - Dallas May 20-21 http://bestpractical.com/training
[rt-users] Missed ticket hits in search
[ Yes, I know RT 3.8 is EOL :) ] One of our RT instances is still running RT 3.8.17. A colleague reported, and I confirmed carefully with my own eyes, that a search is missing tickets it should be finding. Content LIKE '34.100.108.26' AND Created '2014-04-01' AND Queue = 'HJLAM' This returns 2 results. There are at least 2 other tickets that match all 3 items in the search criteria and are not part of the RT search results. Has anyone seen this before? Are there known oddities? The RT instance is using MySQL. Any suggestions? Thanks, Jeff -- RT Training - Dallas May 20-21 http://bestpractical.com/training
[rt-users] Use default global scrips for all queues but one?
Hi all, We 8 queues, all of which make use of the default global scrips. We have a need to set up 1 more queue where we don't want some of the global scrips to run (we want to override some or do away with some entirely). Before I just start flailing at this, I figured I'd ask. I suspect someone has to have wanted this in RT's many years. Is there anything smart in RT to allow this? Do queue-owned scrips take precedence in some way? What's the mechanism? Or am I going to have to (yikes)... make a copy of the current set of global scrips for each of the existing 8 queues, and then 1 more customized copy for the 1 queue? -- RT Training - Dallas May 20-21 http://bestpractical.com/training
Re: [rt-users] Use default global scrips for all queues but one?
4.2 should allow this. We're at 4.2.3. What's in 4.2 that would allow it? Thanks for the reply! -- RT Training - Dallas May 20-21 http://bestpractical.com/training
[rt-users] Posted: Toggle CFs' display based on a control field's value
See: http://requesttracker.wikia.com/wiki/ToggleCFs We had a need to show/hide custom fields depending on the value chosen for a certain control field. That is, if the control field had a certain value, we wanted to show a certain set of custom fields for editing and display. If the control field did not have that value, we wanted to hide those custom fields entirely. This could probably have been done easier in Javascript, but nobody on our team knows Javascript. -- RT Training - Dallas May 20-21 http://bestpractical.com/training
Re: [rt-users] Callback with no TicketObject arg, guidance needed
Alex, thanks again. I promise to write up what I've learned in the last
week (wiki).
I've hit another wall. While I can prove that the callback I am using is
in the right spot and is called...
The following Limit() IS working for Modify.html hits
(EditCustomFields). That is, hide certain fields for editing when a
control flag (CF) is set a certain way.
The following Limit() is NOT working for *Display.html*
(ShowCustomFields) hits. That is, hide certain fields from display when
a control flag (CF) is set a certain way. The problem is, they are still
showing when the control flag is set to not show them.
The code in both callback files is the same.
How might I exclude fields X, Y, Z from $CustomFields in my callback?
# Repeat for each name to exclude:
$CustomFields-Limit(
FIELD= 'Name',
OPERATOR = '!=',
VALUE= 'SomeNameToExclude',
ENTRYAGGREGATOR = 'AND',
);
Here is local/html/Elements/ShowCustomFields/MassageCustomFields (right
spot?) with comments relevant here:
%args
$CustomFields = undef
$ARGS = undef
$Object = undef
/%args
%init
#--
# NOTE: If I enable these 4 short-circuit lines, Display.html does
# in fact only show the 'Scope' custom field as a control/test.
# So a) the callback location seems correct and b) it proves we can
# at least control the contents of $CustomFields.
#
# $CustomFields-Limit(FIELD = 'Name', OPERATOR = '=',
#VALUE = 'Scope', ENTRYAGGREGATOR = 'AND',
#CASESENSITIVE = 0);
# return;
#--
my $ticket = $ARGS{'Object'};
# The fields to hide unless 'Noteworthy = Yes'
my @childfields = (Details, Response, Total emails);
# Default to showing all fields in @childfields
my $showchildren = 1;
# Determine whether to show child fields or not
while (my $cf = $CustomFields-Next) {
my $nam = $cf-Name;
my $val = $ticket-FirstCustomFieldValue($nam);
if ($nam =~ /^noteworthy$/i and $val !~ /^yes$/i) {
$showchildren = 0;
last;
}
}
if (! $showchildren) {
# In theory, this should exclude all fields in @childfields
$CustomFields-Limit(FIELD = 'Name', OPERATOR = '!=', VALUE =
'$_', ENTRYAGGREGATOR = 'AND', CASESENSITIVE = 0) for (@childfields);
}
/%init
%# EOF
--
RT Training - Dallas May 20-21
http://bestpractical.com/training
Re: [rt-users] Callback with no TicketObject arg, guidance needed
Yes -- ish. Note that custom fields can exist on things that aren't tickets, and Elements/EditCustomFields is also used for them. All that's promised is that $Object ISA RT::Record, so you should check $Object-isa(RT::Ticket) before you carry on. Thanks! Follow-up question if I may: How might I exclude fields X, Y, Z from $CustomFields in my callback? I saw in RT-Extension-MandatoryOnTransition some code for INcluding some named fields with 'Limit'. I'm not familiar with Limit (and hopefully its friends). Should I just perldoc these files where 'sub Limit ' is found and start digging around for comprehension? ./lib/RT/Articles.pm ./lib/RT/CustomFieldValues/External.pm ./lib/RT/Tickets.pm ./lib/RT/Links.pm ./lib/RT/Queues.pm ./lib/RT/Users.pm ./lib/RT/SearchBuilder.pm -- RT Training - Dallas May 20-21 http://bestpractical.com/training
[rt-users] Callback with no TicketObject arg, guidance needed
Using RT 4.2.3
Hi all,
This callback in share/html/Elements/EditCustomFields does not pass
$TicketObj (and it could, but was ommitted, perhaps on purpose?):
%INIT
...
$m-callback( %ARGS, CallbackName = 'MassageCustomFields',
CustomFields = $CustomFields );
...
/%INIT
In my MassageCustomFields callback code (per above), I am trying to
determine if a certain CF has a value of /no/i
Given the arguments available to me in the callback, I am not sure how
to do that. All of my work in the past has involved method calls on
$TicketObj.
Debug-printing the contents of %ARGS from inside the MassageCustomFields
callback, I see that ARGS{'Object'} = 'RT::Ticket=HASH(0x7f0b5b540db8)
Just use that as my $TicketObj? Is that sane/safe?
Additionally, if my intention is to ultimately use MassageCustomFields
to modify $CustomFields, isn't that going to fail due to the callback
not being passed a hash _reference_ ?
--
RT Training - Dallas May 20-21
http://bestpractical.com/training
Re: [rt-users] Update.html - Display.html and notifying user, callbacks, etc.
On 3/7/2014 5:36 PM, Kevin Falcone wrote: On Fri, Mar 07, 2014 at 03:39:36PM -0500, Jeff Blaine wrote: This thread is probably going to end up relevant to RT-Extension-MandatoryOnTransition which does not appear to have yet solved this issue. I'm surprised at this statement since MandatoryOnTransition works on Create.html and Update.html but not on Modify.html, but you claim the problem is in Update.html. You are correct. This was a mix-up in my head. Turns out it cost me a lot of time (and some of yours). I apologize. If I am able to come up with something for Modify.html, I will be sure to feed a pull request back to the github repo for RT-Extension-MandatoryOnTransition (not ...MandatoryOnTransaction as I previously worded it). Ticket/Update.html does not redirect to Display.html on submission. It runs and then calls Display.html like any other mason component call using $m-comp (equivalent to . Display.html then redirects to itself if needed. MandatoryOnTransaction uses Update.html/BeforeUpdate to break out of calling Display.html if there was an error. You would use the same callback. If BeforeUpdate sets skip_update and pushes things into the @$results_ref array ref then control will not be dispatched to Display.html and instead Update.html will redisplay with a message at the top for the user. Thank you. This extension and your explanation go places that the wiki examples do not (for those who are not PERL + HTML::Mason devs). -- RT Training London, March 19-20 and Dallas May 20-21 http://bestpractical.com/training
[rt-users] Update.html - Display.html and notifying user, callbacks, etc.
Hi, We're testing a migration from RT 3.8.x (with callbacks, etc) to RT 4.2.3 This thread is probably going to end up relevant to RT-Extension-MandatoryOnTransition which does not appear to have yet solved this issue. I'm a little stuck here and don't see how what I want to do will be possible with the current RT code. I am looking for another set of eyes on the issue if anyone is so generous. The gist is that I am trying to get a message to the user after they have submitted *Ticket/Update.html*. Specifically, they need to be informed that their 'resolve' was rejected due to a Custom Field not having an acceptable value yet. I have no trouble doing this in *Ticket/Modify.html*. That works both with my custom callback code and also completely reimplemented via the RT-Extension-MandatoryOnTransition code (I tested that as well). However, *Ticket/Update.html* redirects to Ticket/Display.html upon submission and is therefore different. Does anyone have any ideas? I've put in a couple of hours now and am still not really sure I understand Update.html and Display.html's interactions with each other (and when). For the time-being I am back to the WWW-circa-1995 behavior, in my callback for Ticket/Update.html, of using Abort(ERROR... hit back in your browser to correct the issue...) Jeff Blaine -- RT Training London, March 19-20 and Dallas May 20-21 http://bestpractical.com/training
Re: [rt-users] A little help with a 4.x callback?
On 2/25/2014 11:26 AM, Kevin Falcone wrote: On Mon, Feb 24, 2014 at 06:10:59PM -0500, Jeff Blaine wrote: I see in 4.2.x (perhaps in 4.0.x) that there is now a ProcessUpdates callback available for use in Ticket/Modify.html I'm reworking some old 3.8 code that would send a message to the end user via (sigh) ... Abort() ... if a certain field had not been set to a value and the user was trying to resolve the ticket. The Ticket/Modify.html callback call reads as: $m-callback( CallbackName = 'ProcessUpdates', TicketObj = $TicketObj, ARGSRef = \%ARGS, Results = \@results ); How might I make use of this @results array in my callback? If you're trying to block the update, ProcessUpdates is too late (all it lets you do is make more changes and push messages back up to the user). Yes, I finally got to that same understanding in the last few hours working on this some more. I'm using 'Default' now. If you want to actively block the user, look at RT-Extension-MandatoryOnTransition. It uses the BeforeUpdate callback Oh, excellent. This will allow us to offload quite a bit of the logic we already have in place and instead focus on the real custom logic we need. in Ticket/Update.html to set skip_create and push onto @results. You can clone that functionality to the Default callback in Ticket/Modify.html. Thanks Kevin. -- RT Training London, March 19-20 and Dallas May 20-21 http://bestpractical.com/training
[rt-users] A little help with a 4.x callback?
I see in 4.2.x (perhaps in 4.0.x) that there is now a ProcessUpdates callback available for use in Ticket/Modify.html I'm reworking some old 3.8 code that would send a message to the end user via (sigh) ... Abort() ... if a certain field had not been set to a value and the user was trying to resolve the ticket. The Ticket/Modify.html callback call reads as: $m-callback( CallbackName = 'ProcessUpdates', TicketObj = $TicketObj, ARGSRef = \%ARGS, Results = \@results ); How might I make use of this @results array in my callback? What should my %ARGS section look like? I looked around, but the wiki is a little lacking in info about how to construct the %ARGS section in the callback file, etc, for those not fluent in Perl past the basics (let alone Mason). -- RT Training London, March 19-20 and Dallas May 20-21 http://bestpractical.com/training
Re: [rt-users] Custom fields export
does anyone knows any easy way to export only custom fields settings, in order to migrate RT 4.2.2 ? Just run duplicate-rt-instance-but-without-tickets.pl :( There is no easy way. Get out your database scalpel. -- RT Training London, March 19-20 and Dallas May 20-21 http://bestpractical.com/training
Re: [rt-users] Server closed connection without sending any data back (500)
If anyone has any ideas how I might go about making RT more informative about what is going on in this failure, I would love to hear them. On 1/9/2014 4:01 PM, Jeff Blaine wrote: I cannot seem to get the RT CLI to work. We are using 4.0.18. Any ideas as to what is going on? The Apache httpd logs show nothing of any use. If I use the web UI (which works fine, and we've been using it for 10+ years now), I can logout + login with the following username and password fine. The user has full create privs for the foo-help queue (it is our day-to-day queue we use all the time). == [rt-tickets:~]$ cat .rtrc server https://rt-tickets.our.org/ user jbla...@our.org passwd mypassword queue foo-help debug 3 [rt-tickets:~]$ == == [rt-tickets:~]$ /rt/bin/rt create -o -t ticket set subject=foo POST https://rt-tickets.our.org//REST/1.0/show Content-Length: 281 Content-Type: multipart/form-data; boundary=xYzZY --xYzZY Content-Disposition: form-data; name=format l --xYzZY Content-Disposition: form-data; name=id ticket/new --xYzZY Content-Disposition: form-data; name=user jbla...@our.org --xYzZY Content-Disposition: form-data; name=pass mypassword --xYzZY-- 500 Server closed connection without sending any data back Content-Type: text/plain Client-Date: Thu, 09 Jan 2014 20:54:30 GMT Client-Warning: Internal response 500 Server closed connection without sending any data back rt: Server error: Server closed connection without sending any data back (500) ==
Re: [rt-users] Server closed connection without sending any data back (500)
On 1/15/2014 11:22 AM, Kevin Falcone wrote: On Wed, Jan 15, 2014 at 09:23:41AM -0500, Jeff Blaine wrote: If anyone has any ideas how I might go about making RT more informative about what is going on in this failure, I would love to hear them. You showed an Internal Server Error 500 Server closed connection without sending any data back But didn't show the logs created by the server for this error. Nobody can help without logs of what happened on the server. That's because there is no 500 error information logged to /var/log/httpd/*log by Apache when I make this request via the RT CLI tool using the settings previously shown in .rtrc. Honest. In fact, there is zero trace of the IP address in all of the Apache logs, even using 'LogLevel debug' for Apache. I can request /sdfkdaslfdskldsfkla and see a 404 just fine and all other normal info is in those logs. It's standard Apache 2.x on RHEL 6. Does the RT CLI tool *support* SSL/TLS 'server' settings?
Re: [rt-users] Server closed connection without sending any data back (500)
Solved: unset http_proxy unset https_proxy *sigh* On 1/15/2014 3:08 PM, Jeff Blaine wrote: On 1/15/2014 11:22 AM, Kevin Falcone wrote: On Wed, Jan 15, 2014 at 09:23:41AM -0500, Jeff Blaine wrote: If anyone has any ideas how I might go about making RT more informative about what is going on in this failure, I would love to hear them. You showed an Internal Server Error 500 Server closed connection without sending any data back But didn't show the logs created by the server for this error. Nobody can help without logs of what happened on the server. That's because there is no 500 error information logged to /var/log/httpd/*log by Apache when I make this request via the RT CLI tool using the settings previously shown in .rtrc. Honest. In fact, there is zero trace of the IP address in all of the Apache logs, even using 'LogLevel debug' for Apache. I can request /sdfkdaslfdskldsfkla and see a 404 just fine and all other normal info is in those logs. It's standard Apache 2.x on RHEL 6. Does the RT CLI tool *support* SSL/TLS 'server' settings?
[rt-users] Server closed connection without sending any data back (500)
I cannot seem to get the RT CLI to work. We are using 4.0.18. Any ideas as to what is going on? The Apache httpd logs show nothing of any use. If I use the web UI (which works fine, and we've been using it for 10+ years now), I can logout + login with the following username and password fine. The user has full create privs for the foo-help queue (it is our day-to-day queue we use all the time). == [rt-tickets:~]$ cat .rtrc server https://rt-tickets.our.org/ user jbla...@our.org passwd mypassword queue foo-help debug 3 [rt-tickets:~]$ == == [rt-tickets:~]$ /rt/bin/rt create -o -t ticket set subject=foo POST https://rt-tickets.our.org//REST/1.0/show Content-Length: 281 Content-Type: multipart/form-data; boundary=xYzZY --xYzZY Content-Disposition: form-data; name=format l --xYzZY Content-Disposition: form-data; name=id ticket/new --xYzZY Content-Disposition: form-data; name=user jbla...@our.org --xYzZY Content-Disposition: form-data; name=pass mypassword --xYzZY-- 500 Server closed connection without sending any data back Content-Type: text/plain Client-Date: Thu, 09 Jan 2014 20:54:30 GMT Client-Warning: Internal response 500 Server closed connection without sending any data back rt: Server error: Server closed connection without sending any data back (500) ==
[rt-users] callback need to load a ticket's CF value from DB, not web form
In my callback for Modify.html, I need to know what the value
of a custom field was set to even if it was unchanged on this
particular form submission.
What I am seeing right now is that if the value was not changed
on this form submission, the value shows in the callback as
an empty string, so I need a function/method that I don't know
about yet, hence this call for help.
Modify.html/jblaine
my $x = $ticket-LoadCustomFieldByIdentifier('Foofield');
my $xx = $ticket-CustomFieldValuesAsString('Foofield');
Both 'x' and 'xx' log as empty strings when I modify a ticket
without touching Foofield (Foofield DOES have a value that was
set previously).
So, cool, fine, Foofield was not modified in this transaction.
I still need to know what its value is from the database in
order to perform some logic if someone is trying to resolve this
ticket.
How do I load Foofield's value from the database?
Re: [rt-users] Validation of Custom Fields with Custom Code
I would like to validate a custom fields with my own custom code (RT 4.0.7).
I saw the code in html/Elements/ValidateCustomFields, but there is no
callback there to implement additional code.
I'm able to write an error message besides my Custom Field (ID 2) in a
Callback Modify.html/Default:
$m-notes( ('InvalidField-2') = $my_msg );
But how can I refuse the change of the custom field?
Steffi, I look forward to someone shedding some light on this too.
Right now we are using the Abort() call and requiring the
user to use the browser back button. It's very ... 1995.
*I* am not aware of any clean way to do what would be desired
by anyone wanting to do something like this:
Evaluate custom logic at Modify.html submission (and
Update.html!), take no action and redisplay the form
with error message if the logic requirements are not met.
[rt-users] Mandatory custom field
Hi all, I am little confused about RT's interpretation of Mandatory as it relates to Custom Fields. I have a CF's validation set to the canned regexp (?#Mandatory) The CF in my test ticket has pointedly been left with no value. Aside from a meager red warning on the page indicating it's not set properly, all other actions on the ticket work just fine. I can even resolve the ticket while the CF is unset ((no value)) So Mandatory how? When does that apply exactly? Does it simply mean, as I seem to be finding, do whatever you want anyway, but we're going to put red text near your offending field ?
Re: [rt-users] REMOTE_USER, external auth, and email mismatching
On 5/15/2013 12:17 AM, Philip Brown wrote: Err.. thanks, but that's not what I'm looking for. For one thing, even if I got permission to do that (which I wont), we have 40,000 users in ldap. I dont actually WANT all of them in the rt database. particularly since we have a 15,000 user/year churn rate. I'm never quite sure why people do that (*import* then-stale data from LDAP/AD which is the dynamic source of truth for the information on the network). There must be some use case that I am just not aware of, but I definitely don't get it. it's kinda odd that I cant seem to google any sample RT_Config.pm files for this Well, here's where I asked the community to share what works: http://lists.bestpractical.com/pipermail/rt-users/2012-February/075085.html And here's 1 (of 2) contributions I got and submitted, still sitting in a github pull request that has not been merged after 1+ years: https://github.com/bestpractical/rt-authen-externalauth/pull/2 That was the water testing and end of my contributions to RT, as a failed experiment. I didn't bother submitting the 2nd config contribution I got based on the above. Here's someone else's code contribution from 1+ years ago to allow secure LDAP: https://github.com/bestpractical/rt-authen-externalauth/pull/1 *crickets* Contributing elsewhere. -- RT Training in Seattle, June 19-20: http://bestpractical.com/training
Re: [rt-users] What sort of site authentication does RT use?
We're attempting to gain access to RT to index it with a third party application over https and we aren't sure what the type of authentication the secure RT site employs when we get to the login screen, Is it Basic HTTP Authentication, NTLM Authentication, Form-Based Authentication? The default RT installation is form-based.
Re: [rt-users] Spreadsheets/Export missing Custom Fields
On 2/21/2013 2:28 PM, Rolf Krogstad wrote: I have just upgraded from RT version 3.6.6 to version 4.0.10. One function the users utilize frequently is to Search for Tickets with specific criteria and then Export them to a Spreadsheet. In the old version they would get the ticket including all the custom fields. After the upgrade to 4.0.10, we now get only the fields that are showing on the Search Results windows. Any ideas? I believe the export will not show the CFs unless they are selected for *display* at the bottom of the search form. -- RT training in Amsterdam, March 20-21: http://bestpractical.com/services/training.html Help improve RT by taking our user survey: https://www.surveymonkey.com/s/N23JW9T
[rt-users] Quietly comment on a ticket?
Is there a way to quietly update a ticket with a note without having to uncheck each one of our ~30 Bcc people at the bottom of the 'Comment' page? Short of their being a way, an Uncheck all would be very welcome in a future release.
Re: [rt-users] Where are comments and correspondence held?
On 11/5/2012 4:54 PM, Max McGrath wrote: Running RT 4.0.6. Curious where the content of *comments* and/or *correspondence* resides in the database? In the transactions table, as I recall. We're hiring! http://bestpractical.com/jobs
Re: [rt-users] Scrips being executed twice for a simple CF update
On 9/14/2012 1:40 PM, Kevin Falcone wrote: On Mon, Sep 10, 2012 at 04:57:33PM -0400, Jeff Blaine wrote: Upon modification of one of our custom fields (text area), we want to parse the data, perform a translation on the original data, and store the results in the custom field. With loads of RT::Logger debugging in place, everything appears to be working fine except for the for the fact that the Scrip is being called twice, and it's performing the change ... then UNDOING IT: I suspect you're running into the processing order of updates on the Basics page. You can try making the Scrip TransactionBatch and see if that resolves it. For the record, that seems to have worked. Thanks Kevin Jeff Keep in mind that TransactionCreate scrips don't see all the updates done on that page at once, it sees them piecemeal. -kevin Final RT training for 2012 in Atlanta, GA - October 23 24 http://bestpractical.com/training We're hiring! http://bestpractical.com/jobs
[rt-users] Scrips being executed twice for a simple CF update
Upon modification of one of our custom fields (text area), we want to parse the data, perform a translation on the original data, and store the results in the custom field. Simplified/contrived example: INPUT: Basics: OurField: 4 5 8 9 32 90 [SAVE]*click* RESULT: Basics: OurField: X Y A B Z O We're obviously using a Scrip to do this field processing. It's not working out too hot, so I *must* be doing something wrong. === # Name: ParseAndWrite_Ourfield # # User Defined # User Defined # Blank template # TransactionCreate Custom condition: if field changed return 1 Preparation: return 1 Clean-up: parse ourfield initialize final_string to for each item found: perform lookup append lookup result to final_string set OurField value to final_string return 1 === With loads of RT::Logger debugging in place, everything appears to be working fine except for the for the fact that the Scrip is being called twice, and it's performing the change ... then UNDOING IT: Sep 10 16:24:39 rtdev2 RT: START SCRIP processing OurField Sep 10 16:24:39 rtdev2 RT: OurField changed: OLD '' NEW '26 34' Sep 10 16:24:39 rtdev2 RT: OurField: In main chunk Sep 10 16:24:39 rtdev2 RT: OurField: 2 values found Sep 10 16:24:39 rtdev2 RT: OurField: Processing: '26' Sep 10 16:24:39 rtdev2 RT: OurField: Processing: '34' Sep 10 16:24:39 rtdev2 RT: OurField: Final string: 'X Y' Sep 10 16:24:39 rtdev2 RT: OurField reached final return 1 Sep 10 16:24:39 rtdev2 RT: START SCRIP processing OurField Sep 10 16:24:39 rtdev2 RT: OurField changed: OLD 'X Y' NEW '' Sep 10 16:24:39 rtdev2 RT: OurField: In main chunk Sep 10 16:24:39 rtdev2 RT: OurField: 0 values found Sep 10 16:24:39 rtdev2 RT: OurField: Final string empty, so clearing out OurField entirely Sep 10 16:24:39 rtdev2 RT: OurField reached final return 1
[rt-users] OnCreate insert a comment IN the ticket?
I'm lost in perldoc lib/RT/*.pm land, confused about how/where to create a comment in a ticket at creation-time. We have a need to put a banner, if you will, as the first comment in every ticket of a particular queue. So, a Scrip that performs a comment action on the ticket. The following fails to do anything. Should I be trying to use $self-TransactionObj-SetData() instead... or...? Scrip: Condition: On Create Action: User defined Template: - Stage: TransactionCreate condition: return 1; action preparation: my $ticket = $self-TicketObj; $ticket-Comment(content = blah!); cleanup: return 1;
[rt-users] Make current queue more obvious (ideally as part of % $title %)
We have a need to make the current working queue more obvious in the web UI. I've had a look, and my first choice desire would be to get html/Elements/PageLayout to include the queue name in $title: ... div id=header h1% $title %/h1 ... Looking at html/Elements/PageLayout's %ARGS, I see that $title = $m-callers(-1)-path How $m-callers(-1)-path gets converted into, say, #5103: my test ticket, is beyond me :( Has anyone already done something like this, and would you be willing to share your solution? Barring that, any leads on the right path to head down would be welcome. I really don't fully comprehend all of the specifics of all of the pieces of RT internals, such as what is available in what scope, how $m works (which I assume is something to do with Mason), etc...
Re: [rt-users] Search results showing a bunch of HTML
On 6/11/2012 5:44 PM, Thomas Sibley wrote: On 06/11/2012 05:11 PM, Jeff Blaine wrote: I've not seen this reported. Any ideas? Performing a search is displaying (showing, not rendering) a bunch of HTML *after the search results* in RT 3.8.12 and RT 3.8.13 (those are the two I know of at least). We can't diagnose this without more details and an example. Well, it's basically exactly as described. Can you tell me what more I can provide? Select Tickets at left to start a search. Do a search. The results appear. There is a pile of visible HTML shown after the results table, starting right next to the [CHANGE] button. Would you like a screenshot?
[rt-users] Search results showing a bunch of HTML
I've not seen this reported. Any ideas? Performing a search is displaying (showing, not rendering) a bunch of HTML *after the search results* in RT 3.8.12 and RT 3.8.13 (those are the two I know of at least).
Re: [rt-users] Scrip/template Notify parent ticket owner on resolve?
On 6/5/2012 6:40 PM, Kenneth Crocker wrote:
Create a Template called Notify Parents on Resolve.
The Template should have code similar to this (loop thru all parents and
add the Emailaddress of the ticket owner to the To: line in the
Template) at the top (test the code to correct my mistakes):
To: {my $parent = $Ticket-MemberOf;); ...
[snip]
Kenn,
That's what I was stumbling toward, but it seemed like I was
perhaps missing a better way to do it. It can be a little
daunting trying to figure out *where* to customize RT before
you even get to the *how* part.
You could have just said, Yes, override the To: with code
in a custom template, but I'll gladly take your much more
detailed response! :)
Thanks.
PS: You got a tip via an Amazon purchase yesterday ;)
Re: [rt-users] Scrip/template Notify parent ticket owner on resolve?
That's a good start to work with. Any leads on the proper way to go about sending mail via a template from inside a scrip would be welcome. Do I just subclass Action::SendEmail, build my message, and commit() it? I don't see a way to reference/use a template for that. Condition: On Resolve Action: User-defined Template: ?? Custom condition code: N/A Custom action preparation code: if ticket has no parents return 1 # otherwise... do stuff, send email Custom action cleanup code: return 1;
Re: [rt-users] Scrip/template Notify parent ticket owner on resolve?
On 6/5/2012 5:29 PM, Thomas Sibley wrote: Thanks Thomas Think about it this way. Make your Condition custom On Resolve + Has Parents Right. Then use a standard Notify* action and set the Template on the scrip. The goal is: On resolve of a child ticket, notify *each parent tickets' owner* that the child was resolved. There is no standard Notify* action for Owners of parent tickets, so I'm confused by what you mean there. You also said the Template. Are you suggesting that I write the logic into a new custom template and force the To: field there? Or did you intend the Template to mean any existing template of my choice. I really don't follow you due to terseness, though I am certainly not looking for anyone to provide the code either.
[rt-users] Scrip/template Notify parent ticket owner on resolve?
Does anyone know of a way to notify the parent ticket's owner on resolve?
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
FWIW, we are seeing the same sendmail problem with 3.8.10 patched with latest patch set. On 5/23/2012 1:06 PM, Thomas Sibley wrote: On 05/23/2012 01:05 PM, Thomas Sibley wrote: That's mighty strange. Can you send us both your non-working and working RT 4.0.6 RT_SiteConfig.pm? Please be sure to redact any passwords contained in them. You can get a redacted version, along with other useful information, by saving the Configuration - Tools - System Configuration page in your browser.
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
On 5/23/2012 1:11 PM, Jeff Blaine wrote: FWIW, we are seeing the same sendmail problem with 3.8.10 patched with latest patch set. [Wed May 23 16:36:25 2012] [crit]: rt-3.8.10-19076-1337790985-875.54825-2...@our.org: Could not send mail with command `/usr/sbin/sendmail -oi -t`: program unexpectedly closed pipe at /apps/rt/bin/../lib/RT/Interface/Email.pm line 451, GEN72 line 1. All defaults, not overridden in RT_SiteConfig.pm Set($MailCommand , 'sendmailpipe'); Set($SendmailArguments , -oi -t); Set($SendmailBounceArguments , '-f '); Set($SendmailPath , /usr/sbin/sendmail); [root@rtsrv1 etc]# uname -a Linux rtsrv1.our.org 2.6.18-308.el5 #1 SMP Fri Jan 27 17:17:51 EST 2012 x86_64 x86_64 x86_64 GNU/Linux [root@rtsrv1 etc]# cat /etc/redhat-release Red Hat Enterprise Linux Server release 5.8 (Tikanga) [root@rtsrv1 etc]# ls -ld /usr/sbin/sendmail lrwxrwxrwx 1 root root 21 Aug 17 2009 /usr/sbin/sendmail - /etc/alternatives/mta [root@rtsrv1 etc]# ls -ld /etc/alternatives/mta lrwxrwxrwx 1 root root 27 Nov 19 2011 /etc/alternatives/mta - /usr/sbin/sendmail.sendmail [root@rtsrv1 etc]#
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
Are you running mod_perl or FastCGI? If fastcgi, mod_fcgid, mod_fastcgi, or something else? mod_perl Can we see a saved copy of the System Configuration page? Mailing to you. What's in your mail logs? Absolutely nothing worthwhile :( Just a bunch of angry users with broken RT mail from a security patch. BTW, as an aside, these are completely conflicting pieces of information: Note that you should remove the '-t' from C$SendmailArguments if you use 'sendmail' rather than 'sendmailpipe' ... These arguments are good for sendmail brand sendmail 8 and newer CSet($SendmailArguments,-oi -t -ODeliveryMode=b -OErrorMode=m);
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
Will also add to the data that our *other* RT instance running 4.0.6 now, on a RHEL 6.2 box, with mod_perl as well... works fine. [ In contrast to our 3.8.10 + security patches on RHEL ] [ 5.8 where web UI - email broke ]
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
On 5/23/2012 2:50 PM, Thomas Sibley wrote: On 05/23/2012 01:54 PM, Jeff Blaine wrote: What's in your mail logs? Absolutely nothing worthwhile :( Can you send us your apache error logs? Honest, there's nothing there to send you. Nothing in error logs at all per-correspondence. Nothing bad in error logs at RT startup. Here's the ssl_access_log info from a reply, which is all I have to show for the attempt: xx.yy.44.251 - - [23/May/2012:15:44:14 -0400] GET /Ticket/Update.html?Action=Respondid=54826 HTTP/1.1 200 14631 xx.yy.44.251 - - [23/May/2012:15:44:19 -0400] POST /Ticket/Update.html HTTP/1.1 302 365 xx.yy.44.251 - - [23/May/2012:15:44:21 -0400] GET /Ticket/Display.html?id=54826results=47c7103ceafa17814c73bfbd05661ca1 HTTP/1.1 200 41286 There's nothing in the mail queue and there's nothing in /var/log/maillog, which we have configured in /etc/syslog.conf as: mail.* -/var/log/maillog Are you running with selinux enabled? Nope. I'll point out, too, for more data ... that before I did the patching yesterday, I did a: mkdir SAVE cp -rp bin lib share local etc SAVE Looking at the new lib/RT/Interface/Email.pm from the patch, and looking at the diff IN the patch file itself, I saw nothing requiring any new code or DB references, so tried dropping the old SAVE/lib/RT/Interface/Email.pm file back in place. It bought me nothing.
Re: [rt-users] emails on ticket updates not being sent in 4.0.6
On 5/23/2012 4:23 PM, Thomas Sibley wrote: On 05/23/2012 04:04 PM, Jeff Blaine wrote: I'll point out, too, for more data ... that before I did the patching yesterday, I did a: mkdir SAVE cp -rp bin lib share local etc SAVE Looking at the new lib/RT/Interface/Email.pm from the patch, and looking at the diff IN the patch file itself, I saw nothing requiring any new code or DB references, so tried dropping the old SAVE/lib/RT/Interface/Email.pm file back in place. It bought me nothing. Did you restart apache after this? We see no reason why this shouldn't work, and it works locally for us. (And if you don't use RT's VERP support, it won't even leave you vulnerable.) Can you try it again, making sure to restart apache, and verify if it does or does not work? I did restart Apache, however I remembering now that we have quite a bit of the correspond scrips disabled (but no, this has not just been a Problem Exists Between Chair and Keyboard situation). Tickling some RT email via a ticket update (owner change) does in fact send email properly with the old Email.pm in place.
Re: [rt-users] RT::Authen::ExternalAuth with AD...
Share the solution? On 4/19/2012 6:46 PM, Glenn Sieb wrote: Thanks to jibsheet Paul in the IRC channel for their help! Best, --Glenn
[rt-users] CF dependencies? Hide some CFs unless another CF has a certain value?
Before I dive into this any further than my current implementation already is, I figured I'd ask: Is there any way to hide and show CFs depending on the value of another CF? We only want an additional list of CFs shown if a certain CF (the control CF) has a certain value. if CF foo has value YES Show CF X, Y, and Z else Do not show CF X, Y, and Z
Re: [rt-users] CF dependencies? Hide some CFs unless another CF has a certain value?
On 3/27/2012 5:05 PM, Ruslan Zakirov wrote: if CF foo has value YES Show CF X, Y, and Z else Do not show CF X, Y, and Z Use custom java script for that. That would be nice. I'm not sure how one would go about that given that the CF table row ids are sequentially named on the Display.html page: tr id=CF-1-ShowRow I can't target that. I see the right thing was done on the edit pages, however... td class=cfname-myfield ...other classes...
Re: [rt-users] RT 3.8.8 to RT 4.0.5 DB Upgrade Warning
On 3/20/2012 5:17 PM, UnixMan wrote: Thanks - do you know how I can rename it ? UPDATE Queues SET Name = '___Approvals' WHERE Name = 'Approvals';
Re: [rt-users] centos OS upgrade broke RT 3.8.2
On 3/19/2012 1:34 PM, testwreq wreq wrote: Hi I upgraded centos from 5.5 to 5.8 release on my development server. I cannot get RT to work. My RT installation is under data/rt3 folder. The document rool is .../share/html; When we try to access the RT host name on the browser, it does not reach the index.html file location in the document root of RT, bur instead tries to open some file and is looking for a filetype to open it with I ran ./rt-test-dependencies --with-modperl1 on both development and production. there are following things missing on both Apache::Request...MISSING CORE missing dependencies: Encode = 2.13...MISSING Encode version 2.13 required--this is only version 2.12 However, the production centos is not updgraded and it is working. I don't get any errors in the error log. Could anyone help us to find what the problem could be? thanks much:) perl -MCPAN -e install Encode Red Hat / CentOS updates nearly always screw up something with the Encode that RT expects. The RH or CentOS one lacks mime_name somewhere.
Re: [rt-users] centos OS upgrade broke RT 3.8.2
On 3/19/2012 1:54 PM, testwreq wreq wrote: What mime_name should I be looking for? None. I guess I was too terse. CentOS upgraded your Perl Encode module for you and the version that RH and CentOS push out breaks RT. The version they push out (which breaks things) is lacking an attribute or function called mime_name. You don't need to know that to fix it, I am only telling you so that if you see mime_name mentioned in an error some time, you will have some context for it. To fix, just do this: perl -MCPAN -e install Encode On Mon, Mar 19, 2012 at 1:47 PM, Jeff Blaine jbla...@kickflop.net mailto:jbla...@kickflop.net wrote: On 3/19/2012 1:34 PM, testwreq wreq wrote: Hi I upgraded centos from 5.5 to 5.8 release on my development server. I cannot get RT to work. My RT installation is under data/rt3 folder. The document rool is .../share/html; When we try to access the RT host name on the browser, it does not reach the index.html file location in the document root of RT, bur instead tries to open some file and is looking for a filetype to open it with I ran ./rt-test-dependencies --with-modperl1 on both development and production. there are following things missing on both Apache::Request...MISSING CORE missing dependencies: Encode = 2.13...MISSING Encode version 2.13 required--this is only version 2.12 However, the production centos is not updgraded and it is working. I don't get any errors in the error log. Could anyone help us to find what the problem could be? thanks much:) perl -MCPAN -e install Encode Red Hat / CentOS updates nearly always screw up something with the Encode that RT expects. The RH or CentOS one lacks mime_name somewhere.
[rt-users] What determines external auth vs. local auth
When using RT-Extension-ExternalAuth, what is the logic for determining whether to authenticate to the external source or via RT's database? Will it always try the external source and then only use local if that was a connection failure or similar? I hope 'root' never tried via external. Is that true?
Re: [rt-users] centos OS upgrade broke RT 3.8.2
On 3/19/2012 3:36 PM, testwreq wreq wrote: Dear Jeff, Unfortunately, it did not fix the problem. Encode was up to date. [root@www2 ~]# perl -MCPAN -e install Encode CPAN: Storable loaded ok Going to read /root/.cpan/Metadata Database was generated on Sun, 18 Mar 2012 23:12:50 GMT Encode is up to date. Did you originally install RT from source? If so, the easiest solution is to visit the original source tree and run 'make fixdeps' and cross your fingers. Alternatively, you can just install a newer Encode (and then move on to Apache::Request). http://search.cpan.org/dist/Encode/ Pick a version, download it, unpackage it, and run perl Makefile.PL make make install [ Or read the README included with the package ] Unfortunately, this is how things go on Linux distributions unless you explicitly exclude key software from being updated. The update changed things that RT wants. The only real solution is to solve it by hand through Perl module upgrades on your own (which may very well then be overwritten again by your package manager in the future!). That's about all I can offer. On Mon, Mar 19, 2012 at 3:03 PM, Jeff Blaine jbla...@kickflop.net mailto:jbla...@kickflop.net wrote: On 3/19/2012 1:54 PM, testwreq wreq wrote: What mime_name should I be looking for? None. I guess I was too terse. CentOS upgraded your Perl Encode module for you and the version that RH and CentOS push out breaks RT. The version they push out (which breaks things) is lacking an attribute or function called mime_name. You don't need to know that to fix it, I am only telling you so that if you see mime_name mentioned in an error some time, you will have some context for it. To fix, just do this: perl -MCPAN -e install Encode On Mon, Mar 19, 2012 at 1:47 PM, Jeff Blaine jbla...@kickflop.net mailto:jbla...@kickflop.net mailto:jbla...@kickflop.net mailto:jbla...@kickflop.net wrote: On 3/19/2012 1:34 PM, testwreq wreq wrote: Hi I upgraded centos from 5.5 to 5.8 release on my development server. I cannot get RT to work. My RT installation is under data/rt3 folder. The document rool is .../share/html; When we try to access the RT host name on the browser, it does not reach the index.html file location in the document root of RT, bur instead tries to open some file and is looking for a filetype to open it with I ran ./rt-test-dependencies --with-modperl1 on both development and production. there are following things missing on both Apache::Request...MISSING CORE missing dependencies: Encode = 2.13...MISSING Encode version 2.13 required--this is only version 2.12 However, the production centos is not updgraded and it is working. I don't get any errors in the error log. Could anyone help us to find what the problem could be? thanks much:) perl -MCPAN -e install Encode Red Hat / CentOS updates nearly always screw up something with the Encode that RT expects. The RH or CentOS one lacks mime_name somewhere.
Re: [rt-users] 4.0.5 FastCGI nightmare
On 2/21/2012 9:43 PM, Kevin Falcone wrote: sudo -u apache plackup --server FCGI /var/www/fcgi/rt-server.fcgi Works fine: FastCGI: manager (pid 21688): initialized FastCGI: server (pid 21691): initialized FastCGI: manager (pid 21688): server (pid 21691) started ^C Apache still just sits there spinning its wheels when hit by browser. [ PS: No firewall in play here either FWIW ] RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] 4.0.5 FastCGI nightmare
Apache LogLevel debug showed that mason_data could not be written to as user 'rt' (it was apache:apache 770) cd mason_data rm -rf * chgrp rt . service httpd restart Win. Well, insofar as using FastCGI via suexec can be considered a win. Thank you for the help. Now then -- before I go bothering, would additional info in docs/web_deployment.pod be welcomed? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] Am I missing a right?
Using RT 4.0.4 If a ticket is owned already, I cannot re-assign it unless I set the owner to Nobody, then assign it. Is this normal behavior? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] Running 3.8.11 and 4.0.5 on the same box?
So, if I build my own separate Perl for RT 4.0.5, how do I make Red Hat's mod_perl 2 use *that* Perl? Seems like a very dark and long rabbit hole... On 2/20/2012 2:18 PM, Jeff Blaine wrote: We want to install 4.0.5 alongside our working 3.8.11 setup, and keep 3.8.11 working while some folks check out 4.0.5. Does anyone know if the deps for 4.0.5 will screw up a running 3.8.10 instance? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] 4.0.5 FastCGI nightmare
Well now. What a total waste of 4 hours so far :( It's a shame this is what web app deployment is in 2012 still. I remember this same old sort of crap from a decade ago. If anyone has any ideas, please share. I'm choosing FastCGI so that I can use my own Perl build, and not Red Hat's built into mod_perl. I've read docs/web_deployment.pod. It includes an Apache snippet, doesn't mention suexec configuration like the wiki does, doesn't mention the suexec DocumentRoot restriction, etc... Not a very useful document here so far. Here's rt-server.fcgi below, copied into the spot where Red Hat's suexec *must* find it located (under the compiled-in DocumentRoot). As you can see, it is owned by the suexec-destination user and group so suexec is happy: [root@rtdev1 httpd]# [root@rtdev1 ~]# cd /var/www [root@rtdev1 www]# ls -ld . drwxr-xr-x 7 root root 4096 Feb 21 20:05 ./ [root@rtdev1 www]# ls -ld fcgi/ dr-xr-xr-x 2 rt rt 4096 Feb 21 20:12 fcgi// [root@rtdev1 www]# ls -ld fcgi/rt-server.fcgi -r-xr-xr-x 1 rt rt 8043 Feb 21 20:12 fcgi/rt-server.fcgi* [root@rtdev1 www]# LoadModule fastcgi_module modules/mod_fastcgi.so FastCgiWrapper /usr/sbin/suexec FastCgiServer /var/www/fcgi/rt-server.fcgi -processes 5 -idle-timeout 300 -user rt -group rt VirtualHost 1xx.xx.xx.xx:443 DocumentRoot /apps/rt4/share/html AddDefaultCharset UTF-8 Alias /NoAuth/images/ /apps/rt4/share/html/NoAuth/images/ ScriptAlias / /var/www/fcgi/rt-server.fcgi/ SuexecUserGroup rt rt Location / Options +ExecCGI AddHandler fastcgi-script fcgi Order allow,deny Allow from all /Location /VirtualHost suexec.log states everything is fine: [2012-02-21 20:29:03]: uid: (rt/rt) gid: (rt/rt) cmd: rt-server.fcgi Apache reports this: [Tue Feb 21 20:33:40 2012] [warn] FastCGI: server /var/www/fcgi/rt-server.fcgi has failed to remain running for 30 seconds given 3 attempts, its restart interval has been backed off to 600 seconds And the browser just spins for a LONG time, then finally throws a 500 Internal Server Error. Testing shows this: [root@rtdev1 httpd]# sudo -u apache /var/www/fcgi/rt-server.fcgi --port WARNING: RT couldn't start up a web server on port . This is often the case if the port is already in use or you're running /var/www/fcgi/rt-server.fcgi as someone other than your system's root user. You may also specify a temporary port with: /var/www/fcgi/rt-server.fcgi --port port Please check your system configuration or choose another port I get the same exact error if I run rt-server.fcgi as root, so it's not a permissions problem. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] Running 3.8.11 and 4.0.5 on the same box?
We want to install 4.0.5 alongside our working 3.8.11 setup, and keep 3.8.11 working while some folks check out 4.0.5. Does anyone know if the deps for 4.0.5 will screw up a running 3.8.10 instance? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] RT-Extension-ExternalAuth questions
If I enable RT-Extension-ExternalAuth, does it take complete control of RT authentication, and therefore I must also specify an ExternalSetting service that maps to our existing 'rt4' database to make use of that? I assume My_Pg is a valid type because Pg is a valid DBI:DBD driver for Perl? I don't see it mentioned anywhere in the README or example RT_SiteConfig.pm Man, if there ever was a need for the wiki to be updated, it's the ExternalAuth page. It covers installation of the module, which is simple already. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] Call for ExternalAuth configurations!
It's RT Community time! I am going to gather ExternalAuth configuration examples (from you!) and add them to the official module's source tree so that people have real-world information to work with when dealing with this beast. Please email me your DATA SANITIZED[1] ExternalAuth configuration and indicate if you would like your name associated with the contribution. *If* you have the time, it would be very nice of you to add worthwhile comments about your configuration file's details. 1. * Remove ALL PASSWORDS! * Change all domain names to example.com! * Don't forget to also change any LDAP 'base' information to dc=example,dc=com, etc. * Change all usernames! Please? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] Bug? TSV data includes HTML br tags
Thanks Kevin.
It's a bug in RT-Extension-SearchResult.XLS after all
(which we use as-is, and also dropped in place, modified,
as ScreenResult.tsv)
RT-Extension-SearchResult doesn't loop over and comma-
separate multi-value CFs like stock Results.tsv
Bug report submitted with fix info:
https://rt.cpan.org/Public/Bug/Display.html?id=74824
On 2/9/2012 5:58 PM, Kevin Falcone wrote:
On Thu, Feb 09, 2012 at 01:36:22PM -0500, Jeff Blaine wrote:
Actually, this is maybe a bug in Mason? If I follow
the code chain:
push @row, ProcessColumnMapValue ($attr,
Arguments = [ $Ticket, 0 ], Escape = 1);
ProcessColumnMapValue() is defined in lib/RT/Interface/Web.pm:
...
return $m-interp-apply_escapes( $value, 'h' ) if $args{'Escape'};
...
So it appears apply_escapes() is leaving behind 'br/' tags?
Has anyone else seen this?
apply_escapes just calls a filter on output. You can use built-in
mason filters or custom ones. RT happens to hook the h one to call
the EscapeUTF8 function in Web.pm, and you can look at what kind of
transform it does.
ProcessColumnMapValue is the method that's used in outputting row data
in ticket listings too, so you can get a sense of where else that code
is run.
-kevin
RT Training Sessions (http://bestpractical.com/services/training.html)
* Boston — March 5 6, 2012
RT Training Sessions (http://bestpractical.com/services/training.html)
* Boston March 5 6, 2012
[rt-users] Bug? TSV data includes HTML br tags
RT 3.8.10 Fields of type Select Multiple (at least) are exporting via Spreadsheet with an HTML br/ tag separating each item selected. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] Bug? TSV data includes HTML br tags
On 2/9/2012 12:39 PM, Jeff Blaine wrote: RT 3.8.10 Fields of type Select Multiple (at least) are exporting via Spreadsheet with an HTML br/ tag separating each item selected. Nevermind. This appears to be a bug in ScreenResults only not Best Practical's code. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] Bug? TSV data includes HTML br tags
Actually, this is maybe a bug in Mason? If I follow
the code chain:
push @row, ProcessColumnMapValue ($attr,
Arguments = [ $Ticket, 0 ], Escape = 1);
ProcessColumnMapValue() is defined in lib/RT/Interface/Web.pm:
...
return $m-interp-apply_escapes( $value, 'h' ) if $args{'Escape'};
...
So it appears apply_escapes() is leaving behind 'br/' tags?
Has anyone else seen this?
RT Training Sessions (http://bestpractical.com/services/training.html)
* Boston March 5 6, 2012
Re: [rt-users] rt-4.0.5 - How to populate a custom field using a web service?
On Feb 6, 2012, at 7:34 AM, Jim Lesinskijim.lesin...@gmail.com wrote: Hi Joe, that's good information but I am looking to be able to use the web service to populate the autocomplete values for one specific RT CustomField within RT. When you are setting up a custom field the screen even says that you can use a web service to populate the field's values, but I am not sure how to do that and I cannot find documentation. Thanks, Jim Lesinski Jim, I believe you've misread the RT CustomField definition screen. I've never used this feature before, but I read and interpret it as follows: It says (emphasis mine): RT can make this *custom field's values into hyperlinks to another service*. Fill in this field with a URL. RT will replace __id__ and __CustomField__ with the record's id and the custom field's value, respectively. Let's say your CF is Employee Number. So, if you set Link values to as: http://service.example.com/lookup?empnum=__CustomField__ If I typed '430' into this CF as a value (as an end user), it would show as a clickable link: http://service.example.com/lookup?empnum=430 Jeff Blaine RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] [rt 4.0.4] Unable to change status from New - Open
You really want all local configs in RT_SiteConfig.pm, not RT_Config.pm. Next time you upgrade, any changes to RT_Config.pm will be blown away. Why doesn't Best Practical just do away this idea? 1. make install put new RT_Config.pm in place make a new 'RT_SiteConfig' help document from it (yes, Site) if there is an RT_SiteConfig.pm in place leave it alone else cp RT_Config.pm RT_SiteConfig.pm 2. Never mention RT_Config.pm anywhere. Mention the help document you made from it. [ Yes, I use RT_SiteConfig.pm, the gigantic WARNING ] [ block in RT_Config.pm was obvious to me ] RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] [rt 4.0.4] Unable to change status from New - Open
Because having an RT_SiteConfig.pm that is the same as RT_Config.pm means that RT can never tell you if you overrode the setting. Sure it can. In those rare needs, the code can just compare the value in RT_Config.pm to the one in RT_SiteConfig.pm. If they're the same, it was not overridden. Also keep in mind that RT can read from 3, 4 or even 5 local config files so you can keep your setting distinct. Having a sparse RT_SiteConfig.pm also makes it trivial as a sysadmin to tell what you've changed locally, especially when looking back through your source control. Isn't that what revision diffs are for? But okay. Just had to throw the idea out there. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] RT 4 Approvals?
Where might I find information on the per-queue setting for Workflow which has a choice called Approval What is the idea? Where is it explained? How does one make use of it? I converted a test queue to use the Approval workflow, submitted a ticket, resolved the ticket, and saw nothing out of the ordinary. It's quite possible this has been around for awhile, but we made the jump from 3.4.4 to 4.0.4. RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
Re: [rt-users] RT 4 Approvals?
The Lifecycle (I'm not sure where Workflow came from) field on a Queue changes which statuses are available in that queue. Changing a random queue from default to approval doesn't turn it into an Approval queue. That Lifecycle is only intended to be used on the ___Approvals queue and there's a bug open to hide it from normal queues to avoid the confusion that it causes. Yes, lifecycle. I don't know where workflow came from. Turns out I can't switch the queue's lifecycle back to 'default' now. RT won't let me (no error, just won't save the change). RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
[rt-users] 3.4.4 -- 4.0.4 and subject tag
Right now, with $rtname set to ATCC and a test queue configured to have its own subject tag (DEMO), our autoreply is showing signs of split-personality disorder: == Subject: [DEMO #35372] my X is broken ... There is no need to reply to this message right now. Your ticket has been assigned an ID of [ATCC #35372]. Please include the string: [ATCC #35372] in the subject line of all future correspondence about this issue. To do so, you may reply to this message. ... == Is this because, perhaps, our old 3.4.4 template was brought into RT 4.0.4 when we upgraded, and it uses $rtname and not something new to reflect the subject tag? If so, what is that? RT Training Sessions (http://bestpractical.com/services/training.html) * Boston March 5 6, 2012
