You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@openmeetings.apache.org by jox joe <jo...@gmail.com> on 2022/06/03 10:52:21 UTC

Re: OM6 testing _ assigning a set of whiteboards to one room

OM 6.2.0 (server RAM: 6 Gbyte)
Scenario:
A teacher/lecturer wants to prepare a complete session of 60 minutes.
He prepares 20 whitebords for this. In this case one whiteboard contains
one slide.
He needs 20 whiteboards with completed content in the given classroom.
The content was prepared with OM and includes text, pictures and videos.
The whiteboards were saved and the list of whiteboards for the given
classroom was generated properly through the OM user interface (see above);
The  list of whitebords is stored in the "room_file" database table.
If we want to put the prepared whitebords on to the wall we need to restart
OM-Tomcat. When we enter the given room the SQL database content
will be copied into a "json" database which is used by the room
subsequently.

This is the result of my testing:

1. If any shape (and its control elements!) is outside the slide boundary
this
scenario is not working: some whitebords are empty.

2. Even if all shapes are placed properly there is a timing problem
during this conversion process: SQL --> json.
Some whitebords were empty.
I modified the source like this and after that it worked roperly:
WbPanel.java:

for (  JSONObject o : wb.list()  )
{

    int b = 0; //inserted line

    while (b < 10000000) //inserted line
    { //inserted line
        b++; //inserted line
    } //inserted line

    arr.put(addFileUrl(cl, wbs.getUid(), o))

    log.debug ( addFileUrl(cl, wbs.getUid(), o) ); // //inserted line

}//END for (  JSONObject o : wb.list()  )

Although this is not the correct solution it shows the nature
of the problem. This was confirmed by the "log.debug" output
in the "openmeetings.log".

Thank you for your attention.
I'm waiting for your feedback/confirmation.

Joe

NOTES:
I improved the OM WB Editor in a way that it is easier to generate content.
Example: shape cloning

OM 6.2.0 is more stable than the previous versions.



jox joe <jo...@gmail.com> ezt írta (időpont: 2022. jan. 11., K, 14:29):

> >did you add all WBs via "Files" section?
> ANSWER: yes
> >have you specified different WB index?
> ANSWER: yes, I assigned unique indices
>
> ADDITIONAL TESTING:
> STEP 1:
> I added 13 whiteboards in the "Files"section as stated above
> Indices:
> 0,1,2,3,4,5,6,7,8,9,10,11,12
>
> STEP 2:
> Stop/start OM-Tomcat
>
> STEP 3:
> Starting the client.
> Missing indeces:
> 4, 8, 10
>
> STEP 4:
> Stop/start OM-Tomcat
>
> STEP 5:
> Starting the client.
> Missing indeces:
> 6,4
>
> STEP 6:
> Stop/start OM-Tomcat
>
> STEP 7:
> Starting the client.
> Missing indeces:
> 5,8,9
>
> STEP 8:
> Stop/start OM-Tomcat
>
> STEP 9:
> Starting the client.
> Missing indeces:
> 12
>
> NOTE:
> Missing index always means that
> the whiteboard is visible but it's empty.
>
> EVALUATION:
> The probability is rather high
> that this is not a database inconsistency
> problem.
>
> ACTION PLAN:
> I'm going to install a clean and latest
> OM on a local Ubuntu server in one or two weeks.
>
> If you can tell me which is the relevant Java module
> maybe I can test it earlier than that.
>
> This was the room table for my test:
> +-----+---------+--------+---------+
> | id  | room_id | wb_idx | file_id |
> +-----+---------+--------+---------+
> | 1   | 2       | 2      | 125     |
> | 2   | 9       | 1      | 126     |
> | 89  | 11      | 0      | 151     |
> | 90  | 11      | 1      | 152     |
> | 91  | 11      | 2      | 153     |
> | 92  | 11      | 3      | 154     |
> | 93  | 11      | 4      | 155     |
> | 94  | 11      | 5      | 156     |
> | 98  | 11      | 9      | 162     |
> | 99  | 11      | 10     | 163     |
> | 100 | 11      | 11     | 165     |
> | 101 | 11      | 12     | 166     |
> | 102 | 11      | 13     | 167     |
> | 103 | 11      | 8      | 161     |
> | 104 | 11      | 14     | 168     |
> | 105 | 11      | 15     | 169     |
> | 106 | 11      | 16     | 170     |
> | 107 | 11      | 17     | 171     |
> | 108 | 11      | 18     | 172     |
> | 109 | 11      | 19     | 173     |
> | 130 | 7       | 0      | 198     |
> | 131 | 7       | 1      | 199     |
> | 132 | 7       | 2      | 201     |
> | 133 | 7       | 3      | 200     |
> | 134 | 7       | 4      | 202     |
> | 135 | 7       | 5      | 203     |
> | 136 | 7       | 6      | 204     |
> | 137 | 7       | 7      | 205     |
> | 138 | 7       | 8      | 206     |
> | 139 | 7       | 9      | 207     |
> | 152 | 8       | 0      | 23      |
> | 170 | 8       | 1      | 7       |
> | 171 | 8       | 2      | 12      |
> | 172 | 8       | 3      | 13      |
> | 173 | 8       | 4      | 212     |
> | 174 | 8       | 5      | 183     |
> | 175 | 8       | 6      | 184     |
> | 176 | 8       | 7      | 209     |
> | 183 | 8       | 8      | 186     |
> | 184 | 8       | 9      | 187     |
> | 185 | 8       | 10     | 188     |
> | 186 | 8       | 11     | 189     |
> | 187 | 8       | 12     | 55      |
> +-----+---------+--------+---------+
>
> The table above contains the initial state
> after restarting OM-Tomcat.
>
> Where do we store the information about
> the set of whiteboards that are actually
> on the wall in a given classroom.
> I mean the the list of active whiteboards
> in a given room.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> jox joe <jo...@gmail.com> ezt írta (időpont: 2022. jan. 10., H, 23:13):
>
>> I'll come back to this issue later.
>>
>> Maxim Solodovnik <so...@gmail.com> ezt írta (időpont: 2022. jan.
>> 9., V, 15:42):
>>
>>> hello
>>>
>>> could you please provide more details on "STEP 4: assign all 20
>>> whiteboards to ONE room (administration/conference rooms)"
>>> did you add all WBs via "Files" section?
>>> have you specified different WB index?
>>>
>>> maybe you can:
>>> - reproduce the issue using "clean" OM
>>> - create JIRA
>>> - attach your DB backup to JIRA so i can check :)
>>>
>>> On Sun, 9 Jan 2022 at 20:12, jox joe <jo...@gmail.com> wrote:
>>> >
>>> > Environment:
>>> > OM 6 (not the latest version)
>>> > mariadb  Ver 15.1 Distrib 10.6.5-MariaDB, for debian-linux-gnu
>>> (x86_64) using readline 5.2
>>> >
>>> > Steps of testing:
>>> > STEP 1: creating 20 new whiteboards (this can be ONE lesson in a
>>> course)
>>> > STEP 2: saving all whiteboards (SAVE button)
>>> > STEP 3: exit the room
>>> > STEP 4: assign all 20 whiteboards to ONE room
>>> (administration/conference rooms)
>>> > STEP 5: exit OM
>>> > STEP 6: restart OM-Tomcat
>>> > STEP 7: enter OM and check the given room which should contain the 20
>>> WBs
>>> >
>>> > My result:
>>> > - some whiteboard contents were missing completely(only the empty WB)
>>> > - some whiteboards were not complete (part of the content was missing)
>>> > - I could put back the missing whiteboard manually, but not always
>>> > - sometimes (error message): "something bad happened, check log,
>>> contact the developers"
>>> >
>>> > Evaluation:
>>> > - the Java module handling this part of the database is suspicious
>>> > (to protect the database consistency "10.6.5-MariaDB" would be able to
>>> give more support /transaction/)
>>> >
>>> > Notes:
>>> > - I tested the same OM 6 on two different Ubuntu 18.4 servers
>>> > - RAM in the first server: 6 Gbyte
>>> > - RAM in the second server: 8 Gbyte
>>> > - the result was the same on both servers
>>> >
>>> > Possible next step:
>>> > - someone in this community could do the same test
>>> > (optimally with the latest "official" OM version)
>>> >
>>> > QUESTIONS:
>>> > What do you think about this test case?
>>> > What is your opinion about creating/managing (lesson) content in OM?
>>> > (digital OM-Whiteboard means you can create your material in advance
>>> > which can include videos, too)
>>> >
>>> > (Reference number of this message: 2022-01-09-13-08-47)
>>> >
>>> >
>>> >
>>>
>>>
>>> --
>>> Best regards,
>>> Maxim
>>>
>>