Known issues

All issues have been addressed. Please let us know if you experience any other problems submitting jobs.

Model Angelo

If your map is not of sufficient quality, there are thousands of segments built by Model Angelo into the map. This large number of segments will cause COSMIC2 to hang.

February 6, 2025

We have received CPU credits to resume CPU job submission.

CPU-based jobs

We are waiting for additional CPU credits to run the following jobs:

  • HHBlits
  • csparc2star
  • LocSpiral, LocBfactor, LocOccupancy
  • CryoEF

We will update this page when we receive additional credits.

January 20, 2025

We are seeing errors related to AlphaFold2, DeepEMhancer, and HHBlits related to job submission. When the user tries to input parameters, there will be a blank set of parameters or blank page.

Please check back here to see when it will be resolved.

December 23, 2024

We are seeing issues related to ColabFold, AlphaFold2, and DeepEMhancer with messages saying:

“Warning, your current total storage size () is at our maximum limit of . Please delete at least as soon as you can.”

“You must enter a value for “maxmsa_”

For Model Angelo, we are seeing the message:

“Mon Jan 06 11:48:11 PST 2025 > COMMANDRENDERING : ERROR : COSMIC2-JOB-MODELANGELO-XXXXXXX: java.lang.RuntimeException: java.lang.NullPointerException: Cannot invoke “java.lang.reflect.Method.invoke(Object, Object[])” because “com.sun.xml.bind.v2.runtime.reflect.opt.Injector.defineClass” is null

“Error uploading: Processing of multipart/form-data request failed. No space left on device”

We are working to fix this and apologize for the inconvenience.

These issues have been resolved January 20, 2025.

Model Angelo

If your map is not of sufficient quality, there are thousands of segments built by Model Angelo into the map. This large number of segments will cause COSMIC2 to hang.

December 8, 2024

We are waiting for additional CPU credits to run the following jobs:

  • HHBlits
  • csparc2star
  • LocSpiral, LocBfactor, LocOccupancy
  • CryoEF

We will update this page when we receive additional credits.

April 2, 2024

GPU job submissions have resumed.

March 31, 2024

We are waiting for new GPU allocations to take effect on the COSMIC2 gateway, which start on April 1st. We have paused GPU submissions at this time. We apologize for any inconvenience. Please check back here for updates.

January 11, 2024

We have run into a job submission allocation error for the following tools. We have paused jobs that use GPU resources. We hope this will be resolved by January 18th.

  • AlphaFold2
  • ColabFold
  • ModelAngelo
  • ESMFold

October 17, 2023

We are experiencing a redirect issue when users try to reach https://cosmic2.sdsc.edu. While we address this issue, you can access the portal directly by typing: https://cosmic2.sdsc.edu:8443/gateway/home.action.

We are experiencing issues where job outputs are not loading the results saying:

Wed Oct 18 11:03:36 PDT 2023 > LOAD_RESULTS : ERROR : COSMIC2-JOB-COLABFOLD-C472540D5427427CB0DD526E55844EE6 : Error retrieving results for COSMIC2-JOB-COLABFOLD-C472540D5427427CB0DD526E55844EE6. Up to 40 automatic retries will be made.

Please email us and we can retrieve the results for you.

September 12, 2023

csparc2star.py – we are experiencing a library dependency issue that we are currently addressing. Tool is nonfunctional.

July 15, 2023

We have paused GPU job submissions while we wait for additional computing credits from ACCESS. We apologize for the inconvenience – the gateway continues to grow exponentially, making managing computing credits challenging.

July 10, 2023

  • We are experiencing login issues due to an issue on the backend of the webserver.

June 1, 2023

  • Job submission issues: Jobs are not submitting due to a high volume of users. We are working on a fix at the moment to address the issue.

May 25, 2023

  • AlphaFold2
    • Plots are not working due to a plotting error. We will resolve this during the week of May 29th.
    • Multimer is generating all models during recycling steps and taking longer than expected
    • Solution: Run jobs with ColabFold
  • Model-Angelo
    • Cannot run with RNA only. There must also be a protein sequence included.
  • cryoDRGN
    • Currently has a GPU-related issue that prevents GPU detection. We are working on this issue.

April 26, 2023

We updated the backend of COSMIC2, which means that the URL cosmic2.sdsc.edu temporarily isn’t working. Instead, you need to use: https://cosmic2.sdsc.edu:8443/gateway

April 7, 2023

Jobs are not submitting to the cluster and remain in ‘queued’ state. We are looking into this.

April 6, 2023

Job submissions resumed.

April 1, 2023

Job submissions paused. We are dealing with an administrative issue with our computing allocation that should be resolved before April 7, 2023.