avinyc

Members
  • Posts

    61
  • Joined

  • Last visited

Posts posted by avinyc

  1. 23 minutes ago, NexusEFR said:

    The problem of putting in production the api of giphy is that they ask for a video where the environment is shown in production where it is displayed by sending a gif, I do not think that admits a capture of the screen where the api in cal. com... but something will try.
    Regarding openssl keys, you mean to put the same in all these variables (adding new variables to the template unraid logically) as the variables used by CALENDSO_ENCRYPTION_KEY and NEXTAUTH_SECRET ? 
     

    Yeah I don't have giphy running on my instance, so I'm just taking guesses here on how to help.

     

    For the variables, all of mine are uniquely generated for each one.

     

    Use openssl rand -base64 32 to generate a key and add it under NEXTAUTH_SECRET.

    Use openssl rand -base64 32 to generate a key and add it under CALENDSO_ENCRYPTION_KEY.

     

    And then another 2 times for the following new ones:

     

    Use openssl rand -base64 32 to generate a key and add it under CALCOM_WEBHOOK_SECRET

    Use openssl rand -base64 24 to generate a key and add it under CALCOM_APP_CREDENTIAL_ENCRYPTION_KEY

     

  2. 1 hour ago, NexusEFR said:

     

    Thanks, yes, the case is that the IDS, the Google API Keys if they work and I can connect Google calendars. But the Giphy API does not work, I thought it was precisely the docker in v3.7.11 and that with the new tag would be resolved. The Giphy API is in normal mode, not in production mode because I understand that it is not necessary, besides complicated to send the requirements that they ask. On the other hand, I still have some notices of NextJS, I hit you, you have any similar?

    Did you upgrade the giphy api to production?  See instructions here: https://developers.giphy.com/docs/api

     

    If you generated the secret keys for CALENDSO_ENCRYPTION_KEY and NEXTAUTH_SECRET correctly, maybe try this section I saw in the sample template here:

     

    # - APP CREDENTIAL SYNC ***********************************************************************************
    # Used for self-hosters that are implementing Cal.com into their applications that already have certain integrations
    # Under settings/admin/apps ensure that all app secrets are set the same as the parent application
    # You can use: `openssl rand -base64 32` to generate one
    CALCOM_WEBHOOK_SECRET=""
    # This is the header name that will be used to verify the webhook secret. Should be in lowercase
    CALCOM_WEBHOOK_HEADER_NAME="calcom-webhook-secret"
    CALCOM_CREDENTIAL_SYNC_ENDPOINT=""
    # Key should match on Cal.com and your application
    # must be 24 bytes for AES256 encryption algorithm
    # You can use: `openssl rand -base64 24` to generate one
    CALCOM_APP_CREDENTIAL_ENCRYPTION_KEY=""

     

  3. On 2/28/2024 at 5:49 AM, NexusEFR said:

    Wow great! Works! Adding the environment variable does not give a login error on the page and works without giving prism or DB errors. Only notices of duplicate values continue to appear in the common.json as mentioned by mariushosting on GitHub.

    Thank you for reporting this problem also in unraid, without your help I would not have been able to continue updating the platform. Best regards

    Yep, i switched back to the latest tag and no prisma errors as well with the new variable added to the template.  The duplicate values in common.json are annoying in the logs, but I don't think it is causing any performance issues.  Let me know if you have any difficulty getting giphy or other apps running.

     

  4. 23 hours ago, NexusEFR said:

    Okay, I'll put some more APIs, but I tried to put the Ghipy and it didn't work well, I said I couldn't find it and that's why I thought it was only for paid cal.com licenses (the api was put inside the cal.com web console, not in docker variables, and that's what I did also with Google's,  many of them can be entered into the web console itself). 
    I have the reverse proxy with NPM (Nginx proxy manager) and it works well in bridge mode with many more dockers that I have running (freescout, pingvin-share, projectsend, s-pdf,etc.) but I didn't know what was better to cal.com a custom dedicated network (local network ip range). And I've tried setting up cal.com with custom networking, but the login error still continues. 
    Thanks again.

    I missed a note at the end of that issue, one commenter said a new variable needs to be added now:

     

    DATABASE_DIRECT_URL="postgresql://postgres:@localhost:5450/calendso"

     

    So we need to add this and keep the duplicate DATABASE_URL in our template also.  Looks like this change happened 2 weeks ago, but I haven't tested to see if this allows things to run correctly after v.3.7.11.

     

    • Like 1
  5. 23 hours ago, NexusEFR said:

    Okay, I'll put some more APIs, but I tried to put the Ghipy and it didn't work well, I said I couldn't find it and that's why I thought it was only for paid cal.com licenses (the api was put inside the cal.com web console, not in docker variables, and that's what I did also with Google's,  many of them can be entered into the web console itself). 
    I have the reverse proxy with NPM (Nginx proxy manager) and it works well in bridge mode with many more dockers that I have running (freescout, pingvin-share, projectsend, s-pdf,etc.) but I didn't know what was better to cal.com a custom dedicated network (local network ip range). And I've tried setting up cal.com with custom networking, but the login error still continues. 
    Thanks again.

    I started experiencing issues with the database and searched for the problem. I came across this issue on github:

     

    https://github.com/calcom/docker/issues/329

     

    The recommendation is to switch the image calcom/cal.com:latest with image: calcom/cal.com:v3.7.11 until they can figure out why the container is failing.  You were right to narrow down the problem starting with version v3.7.15 and I would stick to 3.7.11 for the time being.  I downgraded and zero errors in my logs now, so this should be the image to stick with while troubleshooting how to get your container running correctly.

  6. 21 minutes ago, NexusEFR said:

    Thank you so much for sharing your template, by the way, amazing you have also set up all the API variables for payment gateways (stripe, etc.) for Cal.com and video conferencing systems.

    So do you have an enterprise-licensed version of cal.com? Because the APIs only work in the paid version, or at least that's what they say, although you can also put the API keys on the web platform, right?

     

    Another thing, you have the network in direct custom mode, not in bridge right? I have it in bridge mode, it could affect this maybe.  I don't understand why it doesn't work for me, but I appreciate your effort and support.

    Nope, I set up each one of them following the documentation and do not have the enterprise license.  It's running self-hosted and free without any cost.  The only one I couldn't set up was microsoft teams because it requires a work or school account.  Here is the link to the documentation, but setting them up does not require you having a license: https://cal.com/docs/introduction/quick-start/self-hosting/install-apps

     

    My reverse proxy was set up using @SpaceInvaderOne's letsencrypt tutorial:

     

    It is now swag, but the setup should still work: 

     

     

    The instructions have you set up a custom network for the proxy container to work properly, otherwise I don't think it will protect any of the traffic to your server.  Bridge mode will work if you want to access it directly, but if you are exposing this to the world, you wouldn't want that open your server.  If you haven't set up a reverse proxy, I would start there first.

     

  7. 41 minutes ago, NexusEFR said:

    Do you have a reverse proxy on your unraid? Nginx proxy manager maybe and from what version of the unraid CA store? I'm planning to switch to the official jc21 repository instead of jlesage because it doesn't update and they say it's less secure.

    I use swag (formerly LetsEncrypt) as my reverse proxy.  I attached my .conf file, just need to edit the server details and everything else is pretty much default.

    calcom.subdomain.conf

  8. 34 minutes ago, NexusEFR said:

    Yes, I was also trying to use the :latest because I want a new feature of the new versions (the recipient of the meeting invitation has a text box to put a link to their own video conferencing system on the booking page of the sent or shared appointment) but the login module breaks.

    Could you put your template as you have it with the hidden fields of your sensitive data? I've tried everything, with many variables including... Do you have a reverse proxy on your unraid? Nginx proxy manager maybe and from what version of the unraid CA store? I'm planning to switch to the official jc21 repository instead of jlesage because it doesn't update and they say it's less secure.

    I had to redact a lot of the fields, but this is what my calcom template looks like in unRaid:
    Sample calcom template

  9. 3 hours ago, NexusEFR said:

    Well, I'm still figuring things out. The thing is, the GitHub changelog is not the same as the dockerhub changelog and the tags prove it. 

    https://hub.docker.com/r/calcom/cal.com/tags

     

    If you try to pull down the images between V3.7.11 and v3.7.14 they don't exist, because the available images are these and the ones that can be downloaded as upgradable. So this tells me that version V3.7.15 already breaks the docker login of the cal.com app, because in version v3.7.11 everything works. Someone has tried with the latest version to perform the installation, so calcom/cal.com:latest

     

    image.png.a667512fc519fe5b6dfd0445538b2184.png

    I think docker hub isn't showing those versions is because they came out at the same time and building the containers would have been a waste.  I know the latest container works for me, but again there is another component that may be the culprit here.  What version of postgresql are you running?  I am on 15 (specifically 15.6 according to my logs) and have not upgraded to 16 yet.

     

  10. 4 hours ago, NexusEFR said:

    Could you please tell me what release you have installed or when you have installed it because the pace of releases since February 7th when you published your message, has been frenetic. 

     

    https://github.com/calcom/cal.com/compare/v3.7.10...v3.8.0

     

    Thank you, greetings and have a good weekend.

     

    My repository is set as calcom/cal.com:latest

     

    So it's pulling the latest container from here: https://hub.docker.com/r/calcom/cal.com/tags

    I see the versions here don't match the date of the latest pull: https://github.com/calcom/docker/releases

     

    It's not recommended to use the latest tag but I leave it for some of my containers anyway. If I had to take guess based on the last pull it is probably the v3.7.17 image. 

  11. 5 minutes ago, NexusEFR said:

    Okay, it seems that I have managed to bring it up, and when trying to set the first user as an administrator, on the assistant's web interface, it does not proceed to step 2 and starts returning errors in the console like these:

    @calcom/web:start: 15:56:11:916 ERROR[redactError] Error:  {"name":"PrismaClientKnownRequestError","code":"P2021","clientVersion":"5.4.2","meta":{"table":"public.UserPassword"}}
    @calcom/web:start: PrismaClientKnownRequestError: 
    @calcom/web:start: Invalid `prisma.user.create()` invocation:
    @calcom/web:start: 
    @calcom/web:start: 
    @calcom/web:start: The table `public.UserPassword` does not exist in the current database.
    @calcom/web:start:     at Cn.handleRequestError (/calcom/node_modules/@prisma/client/runtime/library.js:123:6817)
    @calcom/web:start:     at Cn.handleAndLogRequestError (/calcom/node_modules/@prisma/client/runtime/library.js:123:6206)
    @calcom/web:start:     at Cn.request (/calcom/node_modules/@prisma/client/runtime/library.js:123:5926)
    @calcom/web:start:     at async l (/calcom/node_modules/@prisma/client/runtime/library.js:128:9968)
    @calcom/web:start:     at async handler (/calcom/apps/web/.next/server/pages/api/auth/setup.js:1:1616)
    @calcom/web:start:     at async /calcom/apps/web/.next/server/chunks/70349.js:1:974
    @calcom/web:start:     at async /calcom/apps/web/.next/server/pages/api/auth/setup.js:1:3920 {
    @calcom/web:start:   code: 'P2021',
    @calcom/web:start:   clientVersion: '5.4.2',
    @calcom/web:start:   meta: { table: 'public.UserPassword' }
    @calcom/web:start: }

     

    And checking the cal.com documentation on Docker installation mentions this error in Prisma that can occur, and it matches the error in my console logs, but I don't know how to modify that parameter or how to reach it in Docker. Does anyone know how to include that space in the user, in an empty JSON object? 

    https://cal.com/docs/introduction/quick-start/self-hosting/docker#troubleshooting

     

    It's telling you there is a missing table in your database, I am guessing the db is not set up correctly or got corrupted.

  12. 44 minutes ago, NexusEFR said:

    Hello everyone, I have been using this template for several months now and recently there have been many updates from both cal.com and postgres. I don't know why, but it started to fail and didn't show the reservations when I logged in.

     

    I had to downgrade to version v.3.7.11 to make it work again. And then I tried newer versions until I couldn't even log in, so my docker became useless. Interestingly, I noticed that the template also disappeared from the CA of unraid (the one mentioned by ibracorp) and now I'm trying to do it manually but I can't seem to get past connection and prisma errors... 

    @calcom/web:start: The provided database string is invalid. Error parsing connection string: invalid port number in database URL. Please refer to the documentation in https://www.prisma.io/docs/reference/database-reference/connection-urls for constructing a correct connection string. In some cases, certain characters must be escaped. Please check the string for any illegal characters.
    @calcom/web:start:     at Cn.handleRequestError (/calcom/node_modules/@prisma/client/runtime/library.js:123:7090)
    @calcom/web:start:     at Cn.handleAndLogRequestError (/calcom/node_modules/@prisma/client/runtime/library.js:123:6206)
    @calcom/web:start:     at Cn.request (/calcom/node_modules/@prisma/client/runtime/library.js:123:5926)
    @calcom/web:start:     at async l (/calcom/node_modules/@prisma/client/runtime/library.js:128:9968)
    @calcom/web:start:     at async getServerSideProps (/calcom/apps/web/.next/server/pages/auth/login.js:1:10204) {
    @calcom/web:start:   clientVersion: '5.4.2',
    @calcom/web:start:   errorCode: undefined
    @calcom/web:start: }
    @calcom/web:start: react-i18next:: You will need to pass in an i18next instance by using initReactI18next
    @calcom/web:start: 13:35:14:306 ERROR[redactError] Error:  {"name":"PrismaClientInitializationError","clientVersion":"5.4.2"}
    @calcom/web:start: PrismaClientInitializationError: 
    @calcom/web:start: Invalid `prisma.user.count()` invocation:

    I had a $ in the password of the PostgreSQL DB and I changed it in case the illegal character was causing issues, but it still gives an error even though I recreated the DB again (same name and user, I admit it).

     

    I'm desperate because it's an incredible application for managing my environment and it's broken in production. Any similar experiences? Is it a problem with postgres? the latest dockers? Do you know if there are solutions for the converted templates to work? Or is it a matter of deleting the entire DB (which I did) and the DB user (which I didn't) and starting from this point? Thanks for the help, greetings!

    I am not qualified to provide actual support, but as another user that has this working on his unraid server I am happy to share whatever I can to get more people running this container.  It is a great application when working.

     

    I reached out to @Squid about why the template was removed from community apps and was told that it lacked support or project tags with applicable urls to the template.  If the developers or anyone that maintains the templates wants to do that, it would return to community apps.

     

    @Tolete provided his XML template which should work for a manual install.  You would need to follow the steps to add it that way and it should run.

     

    As for the prisma error, I found this link which suggests using this command in your console: 

    npx prisma generate

     

    I don't know if you messed with the postgres settings so it might be safer to do a fresh install of a new DB.  Here's a tutorial video that I used and should be enough to get it running again as long as the container install goes smoothly.

     

    Good luck!

  13. Glad you could get it running.  If you need additional variables to add, they are all listed here: https://github.com/calcom/cal.com/blob/main/.env.example

     

    I had to throw in ALLOWED_HOSTNAMES and RESERVED_SUBDOMAINS I think to get the reverse proxy working correctly.  The rest of them are related to the different apps they offer.  The documentation was very clear to get them installed.

  14. 12 minutes ago, Tolete said:

    did app get removed? not seen it in the app store

     

    Strange, I don't see it in community apps either now.  I don't know what happened but the template is still listed in the link I posted previously.  If you are comfortable with setting up the template manually, you can click on "Click Here To Get More Results From DockerHub" and install the official docker image and add the environment varibles yourself.  I can help share details if it's helpful, but it is strange they removed it from CA for no reason.

  15. Now that I finally updated due to CA deprecating anything below 6.12, I thought I was going crazy trying to figure out how to get my dashboard to show 3 columns.  It's annoying but thankfully nothing else is "wrong" with the update.  I tried the browser data reset but it's still not fixing it.  Will keep an eye on this until someone figures it out or a future update automatically does the trick.

  16. On 11/12/2023 at 9:49 AM, garryslash said:

    Did you by a commercial license from them? From my looking at the source, the payment stuff is all under their commercial license, not open source. Upon that, from when I last tested 6 months or so ago, it was not possible to generally run the application using just the open source code, at least without fairly heavy edits.

    No, it works the same self-hosted as if you use the free version directly on their site.  Everything runs smoothly, but yes you don't get the extra features that are included with a paid license.  If you want to try again, let me know what issue you are having to get it running and maybe I can help.

     

  17. Following up to this, and wanted to thank Ibracorp for adding the unraid template for this software to community apps (https://github.com/ibracorp/unraid-templates/blob/master/cal.com/cal.com.xml).  It works perfectly on my server after tweaking the template settings.  It requires following the documentation section for certain app store installs but very straight forward to modify and run (https://cal.com/docs/introduction/quick-start/self-hosting/install-apps).  Setup instructions to get it initially working are here: https://hub.docker.com/r/calcom/cal.com

     

    For reverse proxy, the ip address instead of the container name worked for me.

  18. 1 hour ago, iXNyNe said:

    LSIO tries pretty hard to notify about important changes:

     

    https://github.com/linuxserver/docker-nextcloud#versions (every repo has a section like this)

    https://github.com/linuxserver/docker-nextcloud/releases/tag/27.0.0-ls252 (every release has notes like this, and on GitHub you can subscribe to email notifications when releases happen)

    https://info.linuxserver.io/issues/2023-06-25-nextcloud/ (if there's any big or breaking changes we usually post a notification to our info site with more detailed instructions like this post, the info site also has an RSS feed and posts also are sent to a dedicated notification channel in our discord)

     

    Auto update is definitely not the safest thing to be doing, but we understand how common it is and try to mitigate as much as we can. Nextcloud was a special case, because prior to this update the installation files were stored on your /config volume and mostly untouched by the container without manually running the occ command to upgrade. This however lead to the container dependencies like PHP updating over time with new releases of the container, and the installed version of nextcloud not being updated, and sometimes the installed (outdated) version of nextcloud would not be compatible with the current dependencies in the image. To solve this, we've build the application itself into the container and cherry picked only the files and folders that need to be persistent to remain in the /config volume. In the future, updating the container should automatically update nextcloud itself, along with any dependencies like PHP.

     

    All that being said, LSIO does still recommend pinning a specific version tag such as lscr.io/linuxserver/nextcloud:version-27.0.0 and applying updates manually by simply changing the version at the end. A proper backup and recovery plan is also important.

    Understood, and I do understand the risks with auto-updating, but it wasn't a problem until it became one for me:)  I will follow this advice and use a specific version tag moving forward.

     

    And I also understand there was a way to discover this change if I was properly monitoring the right places, this is a learning experience for me since I didn't even know there was a discord channel as well.  I'm glad this change is happening, and no matter what LSIO could have done, someone like me would have missed it anyway.  I'm just glad it was nothing serious to my server setup.  Thank you for the time responding, have a good day!

     

    • Like 1
  19. 14 hours ago, blaine07 said:

    Oh wow, this is a big deal. I don’t see a tag for 27 that I’m on… I’m just on “latest”m

     

    I am dumb:

    IMG_5728.thumb.jpeg.37c92a7692353ccf897969d71726ead3.jpeg

    Thanks for sharing the link.  

     

    I wish there was a heads up or warning before this happened.  I appreciate the update being automatic with the container now, but I was caught off guard with my installation upgrading by itself.  I usually hold off until all my apps have been properly upgraded to support the latest version before diving in.  Now I have 3 apps that are disabled because my installation jumped ahead.  Again, my fault for making everything auto-update nightly, but now I will turn that off for nextcloud since it requires special attention when upgrading.

  20. On 2/27/2023 at 4:53 PM, avinyc said:

    I'm running into a strange 502 Bad Gateway issue with the latest proxy configuration file in swag (so this may be a swag problem rather than a bitwarden docker one).

     

    Currently using the 2022-09-08 file version and made my setting changes to the newer 2023-02-13 version of bitwarden.subdomain.conf and the only new changes are the Authentik settings (which are commented out) and the following:

     

     location ~ (/bitwarden)?/api {
            include /config/nginx/proxy.conf;
            include /config/nginx/resolver.conf;
            set $upstream_app bitwarden;
            set $upstream_port 80;
            set $upstream_proto http;
            proxy_pass $upstream_proto://$upstream_app:$upstream_port;

        }

     

    Everything is exactly the same except for this section in both files, but running swag using this one returns a bad gateway error and my vault is blank when accessing it from the browser.  Switching back to the previous conf file, everything is back to normal.  Any ideas what's going or if I am missing something else to adjust elsewhere?

    Resolved - set $upstream_app needs to match the container name (in my case bitwardenrs) and I have not been updating all the sections with each updated .conf file.  The addition of the location api commands triggered the bad gateway even though others were not necessary.  All now match and everything works from web, extension and app.

     

    While this is technically a vaultwarden issue, I think it would have been better placed in the swag support thread.

  21. I'm running into a strange 502 Bad Gateway issue with the latest proxy configuration file in swag (so this may be a swag problem rather than a bitwarden docker one).

     

    Currently using the 2022-09-08 file version and made my setting changes to the newer 2023-02-13 version of bitwarden.subdomain.conf and the only new changes are the Authentik settings (which are commented out) and the following:

     

     location ~ (/bitwarden)?/api {
            include /config/nginx/proxy.conf;
            include /config/nginx/resolver.conf;
            set $upstream_app bitwarden;
            set $upstream_port 80;
            set $upstream_proto http;
            proxy_pass $upstream_proto://$upstream_app:$upstream_port;

        }

     

    Everything is exactly the same except for this section in both files, but running swag using this one returns a bad gateway error and my vault is blank when accessing it from the browser.  Switching back to the previous conf file, everything is back to normal.  Any ideas what's going or if I am missing something else to adjust elsewhere?

  22. On 10/31/2022 at 7:34 PM, 007craft said:

     

     I tried switching to using Jitsi meet on the unraid docker templates instead of docker compose.  I used all 4 containers and got it mostly setup, but alas it doesn't work.  Here are my configs. 
    I get it all working but the problem is the video bridge.  It just has errors and then when users join the meeting everybody else just appears muted and cant see webcams to one another.  We need somebody who knows what their doing to give us the settings they used for the jitsi JVB container (the Jitsi video bridge)

    470542520_jitsimeetsettingsd.thumb.png.db5c8f8e954696c749f3f4e71a2c29a4.png

     

    For focus, meet, video, and xmpp A75G repositories, I got the following error in Fix Common Problems: (jitsi/jicofo:latest) has the following comments: Latest tag no longer exists on dockerHub Additionally, this application has been blacklisted from Community Applications for that reason.

     

    I went to the docker hub for one of them and the tags have changed: https://github.com/jitsi/docker-jitsi-meet

     

    The "latest" tag is deprecated and now "stable" is the correct tag.  If anyone manages to get this working, they'll need to make those changes in each of the containers.  It says CA blacklisted them, so not sure if they will bring them back.

  23. On 2/1/2023 at 8:12 AM, DrBobke said:

    Hi everyone, I have been running an own Nextcloud instance for over a year, protected by a reverse proxy SWAG, setup like SpaceInvader one's channel.  Everything has been working great (except for last week, when I also had an issue with SWAG, which luckily was sorted thanks to this forum above here!).

     

    However, I am sharing a lot of photos and personal videos with family and friends and I often get the feedback that they can't download things from my Nextcloud, which is larger in size (it appears that anything below around 2Gb is fine, but above it times out). I did read some comments on the forum, that it would be SWAG which is the bottleneck, but there doesn't seem to be a solution. So I am using Nextcloud, MariaDB, DuckDNS and Swag.  So not sure where the issue lies, but I want to solve this quickly.

     

    Anyone else that has this issue and maybe some kind of explanation why this is happening/how to solve it?

     

    Thanks in advance,

    best regards,

    DrBobke

     

    Look in your proxy configuration file "nextcloud.subdomain.conf"

     

    You can increase the file size limit.  Mine shows as follows "proxy_max_temp_file_size 2048m"

     

    Change it to 4096m if you just want to double it to 4gb, or set any higher number and see if that helps resolve your timeouts.