Invalid login, please try again after upgrade from 3.5.10 to 3.6.8 - moodle

I have created an identical copy of our production instance to my local environment running XAMPP and I am trying to do the following:
After upgrading Moodle from 3.5.10 to 3.6.8 and restoring all plugins etc. I am unable to login. Before upgrading to 3.6.8, I had already performed the following upgrades,
3.4.1 -> 3.4.9
3.4.9 -> 3.5.10
and everything worked fine during those 2 upgrades. I followed exactly the recommended upgrading method as described in moodle's docs.
After successfully upgrading to 3.6.8, I successfully passed the step about the server requirements for the correct php mysql and db versions, I then successfully passed the next step as well for upgrading the database and made sure that all the plugins are fine and had no errors (just some warnings about a few plugins that can be updated to newer version) and after that step finished too, I then got back to the /admin/index.php screen asking me to login. That seemed weird to me because during the previous updates, it never asked me to login after completing the upgrading, but I said, ok. i'll just login. So after trying to login i am constantly getting the message "Invalid login, please try again"
I have tried almost anything here and by anything I mean the following.
First of all, I checked my hashed password in the local database and it is exactly the same with my hashed password in production database.
I then read an article that from moodle 3.1 the logintoken was introduced and since we use the mb2nl theme, I supposed that I just have to add the new input hidden field for this logintoken to the login form but the field was already there and again that was not the issue.
I am at a point where I cannot think of what else might be wrong and I am stuck.
A few more information about our whole setup, we have a Joomla website and our moodle users are actually coming from Joomla, Therefore we use a plugin named Joomdle which connects these 2 CMSs. Our users register to our Joomla website and the Joomdle plugin sends the credentials to moodle and the user is being created to the moodle CMS as well and an entry is being created in the moodle's "user" table. The difference is that moodle generates a different hashed password from the hashed password that joomla does. Anyway, there were no problems at all with logins or anything else since I tried logging in with multiple user accounts in moodle and all worked fine, before upgrading to version 3.6.8.
This is not a Joomdle issue since I already checked this by completely removing Joomdle and this is not the cause. This is probably coming from moodle core but since moodle has NO error reporting or logging at all, I really cannot understand what this is.
To exclude one more matter linked in this question, I already checked and made sure that moodle's password salt is the same both in production and local environments.
I am not sure what is going on here and this is why I desperately need some help from experts.
Solution:
I was able to solve this problem but forgot to come back here and give the solution.
The problem was that, as I said above, we are using the mb2nl theme and the theme's login form did not include the hidden input field to create the logintoken.
On my question above I write somewhere in the middle "I supposed that I just have to add the new input hidden field for this logintoken to the login form but the field was already there and again that was not the issue". I was looking at the wrong place. The login form that I was looking at, was the Moodle's itself login form and not the theme's login form.
I then found the login form in the theme's files and the logintoken hidden input field was indeed missing from there. I added it and then I was able to login normally.

Happened to me today. Executing the cron.php cli script trigered again the upgrade process and solved my problem.

Related

Symfony 3 form not persisting to database

i am currently building subscription form with symfony 3 and all works like it should in development environment. Once i deploy it to production, all works fine at the beginning - data is pushed to database just like it should for the first subscriber. However when i close the website, open it again, complete the form and try to submit it - data is not pushed to DB. I do not get any particular error at that time, it looks like it is trying to push it to database but then something crashes on the background and i get completed for at the end.
That seem to be deployment issue to me but i cannot seem to figure out which one. I was deploying my app with deployer.org
When deploying i was getting this error "Unable to setup correct permissions for writable dirs. You need to configure sudo's sudoers files to not prompt for password, or setup correct permissions manually." I did not give it any special attention as my website did deploy and 1 user of the website did push to DB just fine.
I am using MySql as database
Any assistance is much appriciated.
Finally got this figured out. In my SubscriberDetails Entity I had id set up to be auto generated (e.g. #ORM\GeneratedValue(strategy="AUTO")) and for reasons i am not sure of and would like to find out this was working well on dev but was not working at all on prod. So solution to this was to remove auto generation from entity level and simulate it in controller via:
$query = $em ->createQuery('SELECT MAX(s.id) FROM AppBundle:SubscriberDetails s');
$newSubscriber ->setId($query->getSingleScalarResult() + 1);
That did the trick for me

Drupal site redirects to installation page

I was working on my branch this morning tweaking some CSS when all of a sudden my Drupal site went into a dead mode. I am running multi site mode and the local.settings.php in the individual sites is being ignored on load.
The parent settings.php file is also being ignored on load.
The site redirects to the installation url: http://site.local/install.php
What could have possibly went wrong and how do I go about resolving that?
I finally figured it out. For some reason it had everything to do with the settings in the sites.php The mapping was not 100% correct even though it initially worked. I had to re-map my multi-sites domains in the sites.php and it worked. So if that happens, double check your multi-sites settings. Make sure domain names match and that your database credentials are correct.. [see above]

Oracle Forms App cannot connect to DB (Oracle 10g)

I have a legacy application created in Oracle Forms.
It's a 10g Oracle database, and Forms 6i. Database works OK, I've set up TNS names correctly, tnsping works, I can connect using sqlplus and sql developer.
I think that Forms is also OK, since it gives me a form where the user should input credentials.
The problem is Forms users can't connect. That very first form that is being called has line:
LOGON('user','pass#TNS', false);
this line gives me form_failure, but I don't know what is the problem. Credentials are definitely correct.
If I try like this:
LOGON('user','pass#TNS', true);
then forms gives me an additional generic dialog to login if the former was unsuccessful. That won't let me in also (I've tried a lot of user/pass combos).
I also did some debugging (as much as I know in forms developer) but without success.
I would appreciate very much any help or something that could point me in right direction.
It is usually TNS problem, you have multiple tns files, double check all of them
TNS_ADMIN is not required

Netbeans project integration with different frameworks

We work as a third party with MVC traditional framework with oracle 11g (no hibernate) with login.jsp page to enter the systems
The main company took the
struts2 and spring and hibernate frameworks approach also with oracle 11g and they have thier login.jsp
When we integrated our work together the login.jsp approved the is the last one due to privileges issues (from the main company) although they are too similar but its programmed with hibernate,struts,spring.
There is one problem occurred that their login.jsp page gives an error (below)
java.lang.IllegalArgumentException: Unknown entity: com.uts.entities.appsecurity.Users
The details is too long so i stored it in jsfiddle just for viewing...
The error
Now i should mention when integration i added some extra files .xml from the main company to work with theire framework in addition to their actual work:
1)libraries
2)persistence.xml
3)struts.xml
4)validators.xml
5)applicationContext.xml
Now if i enter the login.jsp i used to use in development it works but not the main company login.jsp we are looking for a solution for this !
the deploy does not give an error and when pressing login apachetomcat output does not give an error , it just redirect to error page on http://localhost:8084/HumanResources_2/presentation/securityAction.action
The main company login
http://localhost:8084/HumanResources_2/presentation/Login.jsp
Our(my) login
http://localhost:8084/HumanResources_2/login.jsp
Another thing the login page of the main company the text appearing as undefined all of them.
i don`t know what causing this it works before integration , note that the main company using Eclipse am ready for any thing i can provide
the problem was a BUG in netbeans
look at the photo there is line i must remove to make it work the selected one
the check when true the above line added automatically am not sure why but it solved it

Lost Microstrategy Administrator Password?

In researching this scenario, all I could find was a suggestion that you contact Microstrategy, send them your metadata DB and ask them for the password.
This CAN'T be the only way.
Is there no way to simply reset the password in some fashion? Looking at the DSSMDUSRACCT table in the metadata DB, this appears to hold (at first glance anyway) a hashed/encrypted password, but I don't believe this is correct, as every entry for every account is the same in the PASSWD field.
Can someone suggest an alternative beyond sending a database to Microstrategy?
There isn't any alternative. We ended up in similar scenario and had Microstrategy reset the password. Password is not easily recoverable from Medadata since it is secured.
I'm not sure how many projects you have in your metadata or if you have access to another user that may have all privileges; but another option would be to create a new database, use MSTR configuration wizard to populate metadata tables, then duplicate the project across project sources into the new metadata that was just created. The Administrator user in this new metadata will have a blank password. If this is not an option then I recommend to open a support ticket (as others have suggested).
One of the things which one can try is to see any command manager script or system manager workflow. Every MicroStrategy environment may have atleast one of the above two things running in their environment. If you check the file, you may see the password for the administrator user. But in case if there is none of these being used then contacting MicroStrategy support would be the right thing to do.
Thanks,
Nikhil.
Benedict's right, there's no way they would keep passwords in plaintext in the metadata. You'll have to send it to them in a support ticket. It can take quite a few days, but they get the job done.
as soon as you get it reset use object manager, create a configuration object package for the administrator account and anytime you forget it, just redeploy the package to reset it.
You could try using the MicroStrategy Object Manager to copy the Administrator user from another project source/server to this environment.
This should do the trick if you have at least one other user with the right administrative privileges.
The other day I wanted to login to one of our applications. However, I forgot my username and password. I tried using the forgot password wizard but had no luck.
The steps I am about to walkthrough will work with most web applications with average logon security measures. It may not work with MicroStrategy, but should work if the input password is not externally verified.
So this is how I got access:
First I logged into my database manager -If you don't have access to
this... run a scan for 'password' on your server root. For Linux I
write sudo grep 'password' -roni or try: db_pass, dbpass,
database_password, etc as other search terms... The -roni will apply various filtering and
options to the search... aka line number, file name and etc. After
getting some results... find a likely file and open it to see if you
have your database details in there.
Login to your database manager. Find the table associated with your account. (This will usually have the name X_users If the password isn't hashed, just take it and login with it... if it is hashed... copy the hashed value to your clipboard.
Change the directory to your chosen applications root.
Run the scan with the search term 'password' or 'md5(' ...this will hopefully bring up a login controller file.
Edit the line within the function that takes your login input. Change the value 'md5($#####)' with '$#####', (aka removing the md5hashing of your input).
Reload the login page. Now, enter your username and hashed password copied from the database and login.
If done correctly you should now be logged into your web app.
Change the password of the account. Now go back to the login controller file and undo the chnages.
This might not be applicable with your app in question, but it's worth a shot! ....and if it isn't appropriate for your scenario... it could still be very helpful for someone else with a similar problem and different application. I just thought there's a small chance it may help someone who's locked out of one of their web applications. Rather than having to reinstall and lose data.
I'd be happy for you to improve on this or give some feedback.