dhis2-devs team mailing list archive
-
dhis2-devs team
-
Mailing list archive
-
Message #39369
Re: Weird cache issue
Ok. I'm sorry the api query was lacking a parameter (should be
api/dataSet.json&fields=name,shortName,displayName) to show all the fields.
DisplayNames are generated (i.e. they're either the translation if one
exists or they're simply the name). If you're not using translations that's
obviously a dead end though.
2015-09-02 12:21 GMT+02:00 Calle Hedberg <calle.hedberg@xxxxxxxxx>:
> Halvdan,
>
> We do not use translations for any South African instance - all English
> (i.e. the translation table has 0 rows)
>
> We are using either 2.19 or 2.20 (mostly 2.19) - actually a slightly
> modified version of 2.19 in most cases. That version contains additional
> code related to DHIS 1.4->DHIS2 conversions etc, but none of that code is
> in any way linked to this problem.
>
> <instance>/api/dataSets gives the result below, or in other words it only
> show names - and all those names are the correct/updated ones.
>
> I'm not sure where "displayNames" would be coming from - I am not aware
> that data sets HAVE a field called "displayName" or similar. In which table
> can I find that?
>
> Regards
>
> On 2 September 2015 at 13:05, Halvdan Grelland <halvdanhg@xxxxxxxxx>
> wrote:
>
>> Could you go to <instance>/api/dataSets and confirm that the names,
>> shortNames and displayNames are correct? The names displayed should be the
>> displayNames; which reminds me: do you have translations for the given
>> dataSets/current locale? If you do they will be displayed, and they will
>> not have been updated.
>>
>> Q:
>> Is the problem true for all objects in the integrity checks, or just
>> DataSets? What about DataElements?
>>
>> Q:
>> Also, what version are you on?
>>
>> Try running ...
>> select * from translation where objectclass like '%DataSet%';
>> ...and inspect the results (might be a large query if you have a lot of
>> dataSets).
>>
>> Also find the uid of one of the affected DataSets and try to run:
>> select * from translation where objectuid like '<insert uid of object
>> here>';
>> .. any hints there?
>>
>> 2015-09-02 10:40 GMT+02:00 Calle Hedberg <calle.hedberg@xxxxxxxxx>:
>>
>>> Jason,
>>>
>>> 1. Yes
>>> 2. Yes, I changed both name and shortname (identical) - but I do not
>>> know if the Data Integrity checks display name or short name (settings =
>>> short name for that instance).
>>> 3. We are not using any caching proxy
>>>
>>> We did restart Tomcat - which should clear the server cache (?) - and I
>>> did clear all my local caches.
>>>
>>> BUT we had some other strange behaviour with regard to dashboard
>>> privileges yesterday, and it was seemingly fixed by clearing the DHIS2
>>> cache MULTIPLE times (not sure how many, but I only cleared the cache one
>>> time...)
>>>
>>> Regards
>>> Calle
>>>
>>> On 2 September 2015 at 10:10, Jason Pickering <
>>> jason.p.pickering@xxxxxxxxx> wrote:
>>>
>>>> So, coming back to Lars's point.
>>>>
>>>> 1) Did you check the "dataset" table to confirm that your intended
>>>> changes were persisted?
>>>>
>>>> 2) Did you alter the dataset "shortname" and is that the same which you
>>>> are seeing in your integrity checks?
>>>>
>>>> 3) You indicated you cleared the DHIS 2 server cache. Can you confirm
>>>> that you have no other upstream caching proxy which might be caching some
>>>> response?
>>>>
>>>> The data integrity checks are a series of SQL (actually generated by
>>>> Hibernate) queries, so as is usually the case, if you make direct database
>>>> manipulations (and do not clear the server cache and possibly your local
>>>> browser cache) you may see inconsistent results. For this type of
>>>> operation, use of the UI is probably better, or alternative with the
>>>> WebAPI.
>>>>
>>>> Regards,
>>>> Jason
>>>>
>>>>
>>>>
>>>> On Tue, Sep 1, 2015 at 11:57 PM, Calle Hedberg <calle.hedberg@xxxxxxxxx
>>>> > wrote:
>>>>
>>>>> Halvdan,
>>>>>
>>>>> Ah, OK, and not - I was not using any queries, I simply updated the
>>>>> Data Set names directly in the table.
>>>>>
>>>>> As I've said - the weird thing is that all those new names immediately
>>>>> showed up in the UI data set list, but the old names are still appearing
>>>>> wherever the Data Integrity process throws up integrity violations
>>>>>
>>>>> Regards
>>>>> Calle
>>>>>
>>>>> On 1 September 2015 at 18:01, Halvdan Grelland <halvdanhg@xxxxxxxxx>
>>>>> wrote:
>>>>>
>>>>>> Calle, I'm pretty sure Jason was referring to whatever you were doing
>>>>>> to update the records (your point #1 in original email) which was not, as
>>>>>> far as I can understand, using the UI or the web-api. I am not familiar
>>>>>> with "pgremote" but am assuming it's some kind of postgres connector
>>>>>> allowing you to run (sql) queries. We'd need to see those queries in order
>>>>>> to figure out what could have gone wrong.
>>>>>>
>>>>>> 2015-09-01 14:50 GMT+02:00 Calle Hedberg <calle.hedberg@xxxxxxxxx>:
>>>>>>
>>>>>>> Jason
>>>>>>>
>>>>>>> I was not executing any custom sql - I'm running the Data
>>>>>>> Administration -> Data Integrity process checks.
>>>>>>>
>>>>>>> My reference to the sql queries being run relates to my ASSUMPTION
>>>>>>> that the Data Integrity checks are a series of SQL queries - but maybe not,
>>>>>>> maybe the integrity checks are a series of web api calls followed by some
>>>>>>> kind of front-end processing? I just don't know, and I'm not a Java
>>>>>>> developer and I don't have the code base on my laptop so I cannot find out
>>>>>>> myself...
>>>>>>>
>>>>>>> Seems my initial post was not clear enough on this - apologies for
>>>>>>> that.
>>>>>>>
>>>>>>> Regards
>>>>>>> Calle
>>>>>>>
>>>>>>> On 1 September 2015 at 14:31, Jason Pickering <
>>>>>>> jason.p.pickering@xxxxxxxxx> wrote:
>>>>>>>
>>>>>>>> Hi Calle,
>>>>>>>> It is very difficult to say much without knowing what SQL it was
>>>>>>>> you executed. In general, use of the API or UI is usually much safer.
>>>>>>>>
>>>>>>>> Do you have any sort of cache in front of your DHIS2 instance?
>>>>>>>>
>>>>>>>> Can you confirm from the database that the names were actually
>>>>>>>> changed?
>>>>>>>>
>>>>>>>> Regards,
>>>>>>>> Jason
>>>>>>>>
>>>>>>>>
>>>>>>>> On Tue, Sep 1, 2015 at 12:49 PM, Calle Hedberg <
>>>>>>>> calle.hedberg@xxxxxxxxx> wrote:
>>>>>>>>
>>>>>>>>> Lars,
>>>>>>>>>
>>>>>>>>> Nope - all data set names and short names are the same.
>>>>>>>>>
>>>>>>>>> It's just weird, and I don't like inexplicable things happening to
>>>>>>>>> a database....
>>>>>>>>>
>>>>>>>>> When you run data integrity - WHERE are those queries fetching
>>>>>>>>> data from? Are the integrity queries running against the database itself,
>>>>>>>>> or some kind of cache repository?
>>>>>>>>>
>>>>>>>>> Regards
>>>>>>>>> Calle
>>>>>>>>>
>>>>>>>>> On 1 September 2015 at 13:00, Lars Helge Øverland <
>>>>>>>>> larshelge@xxxxxxxxx> wrote:
>>>>>>>>>
>>>>>>>>>> Hi Calle,
>>>>>>>>>>
>>>>>>>>>> perhaps you did not update the data set short names?
>>>>>>>>>>
>>>>>>>>>> Lars
>>>>>>>>>>
>>>>>>>>>> On Tue, Sep 1, 2015 at 11:50 AM, Calle Hedberg <
>>>>>>>>>> calle.hedberg@xxxxxxxxx> wrote:
>>>>>>>>>>
>>>>>>>>>>> Hi
>>>>>>>>>>>
>>>>>>>>>>> I've got a baffling problem (presumably related to some kind of
>>>>>>>>>>> caching):
>>>>>>>>>>>
>>>>>>>>>>> 1. yesterday I updated various data records using pgremote,
>>>>>>>>>>> including data set names. We restarted tomcat after the updates.
>>>>>>>>>>>
>>>>>>>>>>> 2. I've also cleared my browser cache, the DHIS2 cache, updated
>>>>>>>>>>> resource tables, etc.
>>>>>>>>>>>
>>>>>>>>>>> 3. When I now open the list of data sets, all the changed/new
>>>>>>>>>>> names are showing as expected.
>>>>>>>>>>>
>>>>>>>>>>> 4. BUT when I run Data Integrity, a few violations related to
>>>>>>>>>>> data sets are coming up AND THE DATA SET NAMES ARE THE OLD ONES.
>>>>>>>>>>>
>>>>>>>>>>> Any ideas what could cause this weird behaviour?
>>>>>>>>>>>
>>>>>>>>>>> Regards
>>>>>>>>>>> Calle
>>>>>>>>>>>
>>>>>>>>>>> *******************************************
>>>>>>>>>>>
>>>>>>>>>>> Calle Hedberg
>>>>>>>>>>>
>>>>>>>>>>> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>>>>>>>>>>>
>>>>>>>>>>> Tel/fax (home): +27-21-685-6472
>>>>>>>>>>>
>>>>>>>>>>> Cell: +27-82-853-5352
>>>>>>>>>>>
>>>>>>>>>>> Iridium SatPhone: +8816-315-19274
>>>>>>>>>>>
>>>>>>>>>>> Email: calle.hedberg@xxxxxxxxx
>>>>>>>>>>>
>>>>>>>>>>> Skype: calle_hedberg
>>>>>>>>>>>
>>>>>>>>>>> *******************************************
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> Mailing list: https://launchpad.net/~dhis2-devs
>>>>>>>>>>> Post to : dhis2-devs@xxxxxxxxxxxxxxxxxxx
>>>>>>>>>>> Unsubscribe : https://launchpad.net/~dhis2-devs
>>>>>>>>>>> More help : https://help.launchpad.net/ListHelp
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> Lars Helge Øverland
>>>>>>>>>> Lead developer, DHIS 2
>>>>>>>>>> University of Oslo
>>>>>>>>>> Skype: larshelgeoverland
>>>>>>>>>> http://www.dhis2.org <https://www.dhis2.org>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>>
>>>>>>>>> *******************************************
>>>>>>>>>
>>>>>>>>> Calle Hedberg
>>>>>>>>>
>>>>>>>>> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>>>>>>>>>
>>>>>>>>> Tel/fax (home): +27-21-685-6472
>>>>>>>>>
>>>>>>>>> Cell: +27-82-853-5352
>>>>>>>>>
>>>>>>>>> Iridium SatPhone: +8816-315-19274
>>>>>>>>>
>>>>>>>>> Email: calle.hedberg@xxxxxxxxx
>>>>>>>>>
>>>>>>>>> Skype: calle_hedberg
>>>>>>>>>
>>>>>>>>> *******************************************
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Mailing list: https://launchpad.net/~dhis2-devs
>>>>>>>>> Post to : dhis2-devs@xxxxxxxxxxxxxxxxxxx
>>>>>>>>> Unsubscribe : https://launchpad.net/~dhis2-devs
>>>>>>>>> More help : https://help.launchpad.net/ListHelp
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Jason P. Pickering
>>>>>>>> email: jason.p.pickering@xxxxxxxxx
>>>>>>>> tel:+46764147049
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>>
>>>>>>> *******************************************
>>>>>>>
>>>>>>> Calle Hedberg
>>>>>>>
>>>>>>> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>>>>>>>
>>>>>>> Tel/fax (home): +27-21-685-6472
>>>>>>>
>>>>>>> Cell: +27-82-853-5352
>>>>>>>
>>>>>>> Iridium SatPhone: +8816-315-19274
>>>>>>>
>>>>>>> Email: calle.hedberg@xxxxxxxxx
>>>>>>>
>>>>>>> Skype: calle_hedberg
>>>>>>>
>>>>>>> *******************************************
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Mailing list: https://launchpad.net/~dhis2-devs
>>>>>>> Post to : dhis2-devs@xxxxxxxxxxxxxxxxxxx
>>>>>>> Unsubscribe : https://launchpad.net/~dhis2-devs
>>>>>>> More help : https://help.launchpad.net/ListHelp
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>>
>>>>> *******************************************
>>>>>
>>>>> Calle Hedberg
>>>>>
>>>>> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>>>>>
>>>>> Tel/fax (home): +27-21-685-6472
>>>>>
>>>>> Cell: +27-82-853-5352
>>>>>
>>>>> Iridium SatPhone: +8816-315-19274
>>>>>
>>>>> Email: calle.hedberg@xxxxxxxxx
>>>>>
>>>>> Skype: calle_hedberg
>>>>>
>>>>> *******************************************
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Jason P. Pickering
>>>> email: jason.p.pickering@xxxxxxxxx
>>>> tel:+46764147049
>>>>
>>>
>>>
>>>
>>> --
>>>
>>> *******************************************
>>>
>>> Calle Hedberg
>>>
>>> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>>>
>>> Tel/fax (home): +27-21-685-6472
>>>
>>> Cell: +27-82-853-5352
>>>
>>> Iridium SatPhone: +8816-315-19119
>>>
>>> Email: calle.hedberg@xxxxxxxxx
>>>
>>> Skype: calle_hedberg
>>>
>>> *******************************************
>>>
>>>
>>
>
>
> --
>
> *******************************************
>
> Calle Hedberg
>
> 46D Alma Road, 7700 Rosebank, SOUTH AFRICA
>
> Tel/fax (home): +27-21-685-6472
>
> Cell: +27-82-853-5352
>
> Iridium SatPhone: +8816-315-19119
>
> Email: calle.hedberg@xxxxxxxxx
>
> Skype: calle_hedberg
>
> *******************************************
>
>
Follow ups
References
-
Weird cache issue
From: Calle Hedberg, 2015-09-01
-
Re: Weird cache issue
From: Lars Helge Øverland, 2015-09-01
-
Re: Weird cache issue
From: Calle Hedberg, 2015-09-01
-
Re: Weird cache issue
From: Jason Pickering, 2015-09-01
-
Re: Weird cache issue
From: Calle Hedberg, 2015-09-01
-
Re: Weird cache issue
From: Halvdan Grelland, 2015-09-01
-
Re: Weird cache issue
From: Calle Hedberg, 2015-09-01
-
Re: Weird cache issue
From: Jason Pickering, 2015-09-02
-
Re: Weird cache issue
From: Calle Hedberg, 2015-09-02
-
Re: Weird cache issue
From: Halvdan Grelland, 2015-09-02
-
Re: Weird cache issue
From: Calle Hedberg, 2015-09-02