Hi, I'm trying to integrate pgweb with an existing website (hosted on aws beanstalk with nodejs, within a load balanced environment). I want to have a button on the website which opens up the "database explorer", bypassing the pgweb login page as I already have those credentials in the nodejs server hosting the website (and in fact, the user might not even know the credentials to the database, only the credentials to the website).

Currently, the only way for bypassing the login page, from what I understand, is by specifying it in the command line when spinning up the pgweb server, right? Which means that I'll need to spin up a server by demand for each user, and on a different port for each user? So this is not a practical solution (especially with regards to configuring all of the port forwarding rules).

Is there any better way for supporting this scenario?

2 potential solutions:

1. Support passing connection bookmarks in the url for an existing pgweb server, and not just on startup. So I'll be able to spin up a single pgweb server in advance. Related to #204 .
2. Ideally, I would want to have pgweb as a library in my existing server, and not to spin up an additional server. So for a specific sub-domain I will forward the request to pgweb (with an option for me to hook to the output before rendering, for example for giving better matching colors for my website).

Thanks.

I am trying to connect to a database running on another machine on the network. When I connect with pgAdmin it works perfectly, so I know that it should work.

I try to run pgweb with the following command:

 pgweb_windows_amd64.exe /url:"postgres://vagrant:[email protected]/mydbname"

Here is the error that it returns:

Connecting to server...
Checking tables...
Error: Database does not have any tables

I turned on connection logging on the server, and here is what I got:

2014-10-29 01:11:30 GMT 54503ec2.7ee9 vagrant mydbname LOG:  statement: SELECT table_name FROM information_schema.tables WHERE table_schema = 'public' ORDER BY table_schema,table_name;
2014-10-29 01:11:30 GMT 54503ec2.7ee9 vagrant mydbname LOG:  could not receive data from client: Connection reset by peer

When I run that query (SELECT table_name FROM information_schema.tables WHERE table_schema = 'public' ORDER BY table_schema,table_name;) in pgAdmin, it returns the list of tables, as expected.

What is causing this?

Is there any other information that can be used to troubleshoot?

So this is semi/totally bullshit, but I figured I'd open the PR for the purpose of further conversation.

The primary problem is (I think) the vendoring of bindata.go. Obviously idomatic go provides no guidance on how to manage an asset pipeline today, and the popular kr https://github.com/kr/heroku-buildpack-go buildpack doesn't know about how to use your Makefile.

Adding a Procfile, app.json and vendoring godeps all seem more-or-less reasonable to me.

Oh, I also stopped checking if there were any tables and crashing if not. pgweb should probably work on empty databases. I mean... why not?

A logo which isn't your picture for the app.json would be cool too, but with this branch you can hit the deploy button and get a working app.

Hello,

Apologies ahead of time if this is not the proper forum but I didn't see any pgweb questions on StackOverflow. I am trying to use pgweb to view the contents of a table with a text column. Anytime I see the results it always displays the results as some numeric value like "41006". All other columns render perfectly. I know the contents are something other than "41006" because the application I am working on renders the contents correctly. I know that the contents of the column are JSON, in case this has any effect.

Thanks, Rex Sheridan

This PR adds the possibility to automatically discover AWS RDS instances and offer them in a dropdown to prefill the host and port fields.

• discovery is disabled by default
• region and credentials can be automatically discovered from EC2 metadata or the usualy AWS SDK config methods (~/.aws/config, env variables, ...)
• in read-only mode the endpoint for RDS clusters returned is the reader endpoint

Hello! This looks terrific. For my use case, I'd love to run this container alongside my postgres container, link them together, and have pgweb available at some port on my URL. Is that an idea that's been had before?

> docker run -p 8082:8081 -e DATABASE_URL='postgres://xxxxx:[email protected]:5444/xxxxx?sslmode=disable' sosedoff/pgweb
Pgweb v0.11.6 (git: 3e4e9c30c947ce1384c49e4257c9a3cc9dc97876) (go: go1.13.7)
Connecting to server...
Connected to  
Checking database objects...
Error: pq: schema "dbms_alert" does not exist

When I launch without DATABASE_URL and then login to the same db http://localhost:8081/api/objects responds with:

{"error":"pq: schema \"dbms_alert\" does not exist","status":400}

My other database tool is Intellij IDEA. I don't see dbms_alert in any part of the db object browser in IDEA.

Hi! I picked up a closed PR #308 for issue #281 and fixed all issues you mentioned. The only problem is that I'm unable to do make text. Make asks password several times and fails. I changed postgres user password, but it didn't help. I guess something is wrong with my setup. Application itself works fine. Still there are not system tests for CSV import, but I'd like to add them only when I'm sure I can run test suite. Can you please elaborate how to set things up?

Hi,

I am trying to get this working under Windows 10 using SSL. However when I make a connection I get:

'sslkey' is not recognized as an internal or external command,
'sslmode' is not recognized as an internal or external command,
'sslrootcert' is not recognized as an internal or external command,

On a Windows system where do I get those binaries?

I have pgweb running in Kubernetes locally with a port forward allowing me to access it.

When I try "Export to SQL" on any random table, it opens up a new tab about:blank. All the other export options do work fine, and open up a popup allowing me to select the download destination.

Since the other functions work fine, I would expect the SQL export to also work.

I'm not sure if this bug is specific to my set up, or pgweb in general

Maybe there is something wrong in the syntax, but this does not work

$ pgweb --ssl disable --url postgres://vc:[email protected]:5432/my_db
Connecting to server...
Error: pq: SSL is not enabled on the server

I'm using 0.3.0, Mac OS X

Frontend code is seriously dated and needs a fresh rewrite, using some minimalistic js framework.

A few things to consider:

• No react
• No unstable frameworks
• No bloatware

This PR adds a "fallback" query to the connection Info() function in pkg/client/client.go. This "fallback" query is invoked when specific managed postgresql hosts deny queries to specific inet_* postgres functions, such as inet_client_addr(), inet_client_port(), inet_server_addr(), and inet_server_port(). The "fallback" query does not include any network information and is generally safe to run on all postgres environments.

This PR closes https://github.com/sosedoff/pgweb/issues/574 by enabling support for bit.io (see screenshots below). We suspect that this PR may also unblock other users of pgweb who are trying to connect pgweb to a postgresql instance with restricted function support (e.g. postgres as a service platforms).

This PR is a rather small delta and should not change any current functionality of pgweb, although if there are any design concerns please let us know.

When connected to Postgres database with hundreds of thousands of tables or with highly complex structures the auto-completion in the SQL editor makes editing real pain because of editor lags for several seconds due to the generation of the auto-completion lists (I guess). Could you please add a switch to turn the auto-completion on/off and an configuration option for it's default setting (on/off)? Thank you.

Further proposals: Also, what's somewhat related is that the initial loading of the database structure (in the left pane) blocks the WUI. It would be nice if this was somewhat asynchronous, just indicating somehow that it's not completely loaded yet and progressing to show what's already scanned but in a way it does not block the rest of the WUI (so that one could connect and start typing SQL immeditately without the need to wait until the left pane loads completely). Thank you.

When pgweb is terminated any new request (or a custom query) will never seem to finish, however the backend is gone and we should print out an error message.