Cannot republish qmd file to Confluence - API Error: 500 - Internal Server Error

[mcutt0523]

Bug description

Publishing a qmd document as a Confluence page for the first time works perfectly - this feature is fantastic and I want my team to use it to publish all of our reproducible analysis. However, when you try to republish a qmd document containing images e.g those created from a ggplot, you get API Error: 500 - Internal Server Error, seemingly because the API is failing to update/overwrite the previous attached images. Republishing to a confluence page, if all existing attachments have been deleted, also works without issue.

ERROR: API Error: 500 - Internal Server Error

Stack trace:
    at ConfluenceClient.handleResponse (file:///Applications/quarto/bin/quarto.js:117390:19)
    at ConfluenceClient.fetchWithAttachment (file:///Applications/quarto/bin/quarto.js:117211:25)
    at eventLoopTick (ext:core/01_core.js:175:7)
    at async ConfluenceClient.createOrUpdateAttachment (file:///Applications/quarto/bin/quarto.js:117360:31)
    at async uploadAttachment (file:///Applications/quarto/bin/quarto.js:117627:32)
    at async Promise.all (index 0)
    at async updateContent (file:///Applications/quarto/bin/quarto.js:117662:45)
    at async doOperation (file:///Applications/quarto/bin/quarto.js:117769:33)
    at async withSpinner (file:///Applications/quarto/bin/quarto.js:84926:16)
    at async doWithSpinner (file:///Applications/quarto/bin/quarto.js:116743:12)

Steps to reproduce

Run quarto publish confluence for the following qmd file, then once it is successfully published to confluence, try to run the same command again to republish the qmd file. This is the point where myself and my entire team receive the API 500 error.

---
title: "Updating Charts Demo"
format: confluence-html
execute: 
  eval: true
  echo: false
  message: false
  warning: false
---

```{r}
#| include: false

library(ggplot2)

# Create sample data
df <- data.frame(
  x = 1:100,
  y = rnorm(100),
  group = rep(letters[1:5], 20),
  size = runif(100, 1, 10)
)

# Store 5 different plots
plot1 <- ggplot(df, aes(x, y)) + geom_point() + labs(caption = paste("Updated at",Sys.time()))
plot2 <- ggplot(df, aes(x, y)) + geom_line() + labs(caption = paste("Updated at",Sys.time()))
plot3 <- ggplot(df, aes(x, y, color = group)) + geom_point() + labs(caption = paste("Updated at",Sys.time()))
plot4 <- ggplot(df, aes(x, y)) + geom_smooth() + labs(caption = paste("Updated at",Sys.time()))
plot5 <- ggplot(df, aes(group)) + geom_bar() + labs(caption = paste("Updated at",Sys.time()))


```

```{r plot1}
plot1

```

```{r plot2}

plot2

```


```{r plot3}

plot3

```


```{r plot4}

plot4

```


```{r plot5}

plot5

```

Actual behavior

The API 500 error means that quarto documents are not republished successfully.

Expected behavior

It should be possible to republish a qmd file with analysis and overwrite the existing attached chart images from the latest version of the analysis. Republishing a qmd file should not return an API 500 error.

Your environment

-IDE: RStudio 2024.12.1+563
-OS: MacOS Sequoia 15.4

Quarto check output

Quarto 1.7.23
[✓] Checking environment information...
      Quarto cache location: /Users/michaelc/Library/Caches/quarto
[✓] Checking versions of quarto binary dependencies...
      Pandoc version 3.6.3: OK
      Dart Sass version 1.85.1: OK
      Deno version 1.46.3: OK
      Typst version 0.13.0: OK
[✓] Checking versions of quarto dependencies......OK
[✓] Checking Quarto installation......OK
      Version: 1.7.23
      Path: /Applications/quarto/bin

[✓] Checking tools....................OK
      TinyTeX: (not installed)
      Chromium: (not installed)

[✓] Checking LaTeX....................OK
      Tex:  (not detected)

[✓] Checking Chrome Headless....................OK
      Using: Chrome found on system
      Path: /Applications/Google Chrome.app/Contents/MacOS/Google Chrome
      Source: MacOS known location

[✓] Checking basic markdown render....OK

[✓] Checking Python 3 installation....OK
      Version: 3.11.5 (Conda)
      Path: /Users/michaelc/anaconda3/bin/python
      Jupyter: 5.7.2
      Kernels: python3

[✓] Checking Jupyter engine render....OK

[✓] Checking R installation...........OK
      Version: 4.3.0
      Path: /Library/Frameworks/R.framework/Resources
      LibPaths:
        - /Users/michaelc/Library/R/arm64/4.3/library
        - /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/library
      knitr: 1.42
      rmarkdown: 2.21

[✓] Checking Knitr engine render......OK

[cscheid]

The API 500 error means that quarto documents are not republished successfully.

500 means "internal server error". Quoting from the MDN page:

If you're a visitor seeing 500 errors on a web page, these issues require investigation by server owners or administrators.

It means an error in the confluence server, not in Quarto. Without additional information, I'm going to close this as it's not our bug something we can act on (EDIT: I said something inaccurate here; I apologize).

[mcutt0523]

Apologies for the delayed reply - thank you for your response. I have tried updating an attachment in the same document using python code, rather than via Quarto. This code enables me to update an attachment in a confluence document successfully:

import requests

# Configuration
base_url = "https://your-domain.atlassian.net/wiki"
page_id = "1234567890"
file_path = "plot1-1.png"
filename = file_path.split("/")[-1]
auth = ("your-email@example.com", "your-api-token")

# Step 1: Get attachment ID
lookup_url = f"{base_url}/rest/api/content/{page_id}/child/attachment?filename={filename}"
resp = requests.get(lookup_url, auth=auth)
attachment_id = resp.json()["results"][0]["id"]

# Step 2: Update attachment content
upload_url = f"{base_url}/rest/api/content/{page_id}/child/attachment/{attachment_id}/data"
with open(file_path, "rb") as f:
    files = {"file": (filename, f)}
    headers = {"X-Atlassian-Token": "no-check"}
    response = requests.post(upload_url, auth=auth, headers=headers, files=files)

if response.status_code == 200:
    print("✅ Attachment updated successfully.")
else:
    print(f"❌ Failed to update attachment: {response.status_code}")
    print(response.text)

Given I've been able to successfully update an attachment using the API outside of Quarto, I feel this does warrant further investigation please!

[cscheid]

@mcutt0523 Sorry, let me say more directly: it's the server returning 500. That means it's not Quarto with an error. It's the confluence server. We really do not have the ability to change how your server responds, and your server is saying "Internal Server Error". I will link to the definition of the 500 error code again: https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status/500

[mcutt0523]

Thank you @cscheid . I understand that it's not Quarto with an error, but I don't receive a server error when I use Quarto publish for the first time on a document. That always works without fail. Specifically updating multiple attachments within a document when republishing via Quarto is what fails consistently. Now that I've shown I can update an attachment in a document using the Confluence API on the same server outside of Quarto, I don't understand what the issue could be and need help finding a solution, whether that is from Quarto or someone else.

What would make the Confluence server return a 500 error when I republish a document and its attachments are updated, vs when I create a document for the first time?

I've done some further testing of this issue morning, and have found that I do not receive a 500 error and can successfully republish a page and update its attachments if the page only has one file (e.g 1 ggplot). So it can work via Quarto, but not for more than 1 attachment...

Can you please confirm whether the example code I shared with you works as expected for you?

[cscheid]

Can you please confirm whether the example code I shared with you works as expected for you?

I really don't know how to say this differently. I'm not suggesting that Quarto is working as expected. I'm suggesting that we cannot act on it if the server doesn't tell us what is going on wrong. The server is saying "sorry, I crashed", and you're giving me some code that works on your server.

Again, the problem here is that the claim you are making in the original post is true, but it's not something we can do much about.

Republishing a qmd file should not return an API 500 error.

If the server had returned something like a 400 (bad request) status code with information about what Quarto is doing wrong, then we could have a clue on how to fix it. I don't have access to your server to know why it's responding in this way, and the error message has no information that is actionable on our part.

[cderv]

It seems Confluence is badly designed and it returns 500 error instead of 400 error 🤦

• https://community.atlassian.com/forums/Confluence-questions/500-Error-when-trying-to-update-content-through-API-rest-api/qaq-p/1891249

So it could be something in how the payload is formatted.

Really hard to know exactly what is the problem, but I'll check how the API call is formatted.

@mcutt0523 a few more details please

• Did you ask for the server log ? Can you contact your IT admin who manage the server to know what this Error 500 is logging for them.

• As it works with Python, I'll compare what we do.

Also please do run again the quarto publish to update, but activate the verbose mode so that we have more information on what quarto is doing. See https://quarto.org/docs/troubleshooting/#verbose-mode

Thank you !