Version 4.7.1 EER issue

Import and Export
1

I had some data in eer format collected with a Falcon 4i in a 200 kV microscope. The data were imported properly into a version 4.0, but failed in a v4.7.1, the latest. The error is the following.

Traceback (most recent call last):
File “cryosparc_master/cryosparc_compute/run.py”, line 129, in cryosparc_master.cryosparc_compute.run.main
File “/cryosparc/cryosparc_worker/cryosparc_compute/jobs/imports/run.py”, line 958, in run_import_movies_or_micrographs
datashape, data_format = read_movie_header(abs_path, extn,
File “/cryosparc/cryosparc_worker/cryosparc_compute/jobs/imports/run.py”, line 785, in read_movie_header
nz, ny, nx = eer.read_eer_shape(abs_path)
File “/cryosparc/cryosparc_worker/cryosparc_compute/blobio/eer.py”, line 15, in read_eer_shape
ioengine.eer_get_inherent_size(fname, xyf)
RuntimeError: Couldn’t open specified EER file

Did any of you ever encounter similar issues? I would appreciate your advice and help. Thanks.

2

Is the data still mounted in the same place? Only time I see this is when a remote disk is unmounted.

3

Yes, the data are local, in a separate drive, not a remote one. I could try to mount it to scratch directory on the same drive as the cryosparc installation directory.

4

Do you have any spaces in the directory structure? That can have odd things happen too.

5

No space is the directory structure. the data are in a drive (/dev/sda) that is about 93% full.

6

Can I do a eer to tiff conversion separately in command line in order to solve this issue?

7

Well, you could, but it wouldn’t solve the apparent underlying issue…

Does the CryoSPARC user have read permissions for the files?

8

Yes, the cryosparc_user does have read permissions.

9

I downgraded to v4.6.0, and the problem still exists.

10

Seems like a malformed wildcard pointing to the data, then, but at this point I’m out of ideas for now…

1 Like
11

It is pretty strange.

12

Is there a format issue? I copied one EER file into the scratch directory. CryoSPARC still coul not open the file.

13

From the system I noticed that app_api_dev is not runing (not installed) in my system (Ubuntu). Would that make a difference in reading eer files?
Thanks.
QX

14

Welcome to the forum @jiangq99920032 .

app_api_dev STOPPED Not started is expected for a v4.7.1 CryoSPARC instance.

Please can you post the output of these commands for the successful (v4.0, if job is still available) and the failed (v4.6 or v4.7) Import Movies jobs

csprojectid=P99 # replace with actual project ID
csjobid=J199 # replace with actual ID of import job
cryosparcm cli "get_job('$csprojectid', '$csjobid', 'job_type', 'version', 'status',  'params_spec', 'errors_run')"
cryosparcm eventlog $csprojectid $csjobid | tail -n 50
15

The EER file format has been tweaked a few times, but I’ve never had CryoSPARC throw incompatibility errors (although the 2K EER from Falcon C needed special handling IIRC).

16

Please see below:

cryosparc_user@jianglabDell7960:/data2/haierData/CS-haier-freezer-test-data$ cryosparcm cli "get_job('$csprojectid', '$csjobid', 'job_type', 'version', 'status',  'params_spec', 'errors_run')"
{'_id': '686bb51f570568d067d28f3e', 'errors_run': [{'message': "Couldn't open specified EER file", 'warning': False}], 'job_type': 'import_movies', 'params_spec': {'accel_kv': {'value': 200}, 'blob_paths': {'value': '/cryosparc/data01/FoilHole*EER.eer'}, 'cs_mm': {'value': 2.7}, 'eer_num_fractions': {'value': 36}, 'eer_upsamp_factor': {'value': 2}, 'gainref_path': {'value': '/cryosparc/data01/20250619_153742_EER_GainReference.gain'}, 'psize_A': {'value': 0.89}, 'skip_header_check': {'value': False}, 'total_dose_e_per_A2': {'value': 40.2}}, 'project_uid': 'P4', 'status': 'failed', 'uid': 'J3', 'version': 'v4.7.1'}
cryosparc_user@jianglabDell7960:/data2/haierData/CS-haier-freezer-test-data$ cryosparcm eventlog $csprojectid $csjobid | tail -n 50
[Wed, 09 Jul 2025 01:45:27 GMT]  License is valid.
[Wed, 09 Jul 2025 01:45:27 GMT]  Launching job on lane default target jianglabDell7960 ...
[Wed, 09 Jul 2025 01:45:27 GMT]  Running job on master node hostname jianglabDell7960
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Job J3 Started
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Master running v4.7.1-cuda12, worker running v4.7.1-cuda12
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Working in directory: /data2/haierData/CS-haier-freezer-test-data/J3
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Running on lane default
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Resources allocated:
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB]   Worker:  jianglabDell7960
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB]   CPU   :  [0]
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB]   GPU   :  []
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB]   RAM   :  [0, 1, 2]
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB]   SSD   :  False
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] --------------------------------------------------------------
[Wed, 09 Jul 2025 01:45:28 GMT] [CPU RAM used: 94 MB] Importing job module for job type import_movies...
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 319 MB] Job ready to run
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 319 MB] ***************************************************************
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 319 MB] Importing movies from /cryosparc/data01/FoilHole*EER.eer
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 319 MB] Importing 3229 files
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 320 MB] Import paths were unique at level -1
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 320 MB] Importing 3230 files
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 320 MB] Reading header for each exposure...
[Wed, 09 Jul 2025 01:45:33 GMT] [CPU RAM used: 321 MB] Traceback (most recent call last):
  File "cryosparc_master/cryosparc_compute/run.py", line 129, in cryosparc_master.cryosparc_compute.run.main
  File "/cryosparc/cryosparc_worker/cryosparc_compute/jobs/imports/run.py", line 958, in run_import_movies_or_micrographs
    datashape, data_format = read_movie_header(abs_path, extn,
  File "/cryosparc/cryosparc_worker/cryosparc_compute/jobs/imports/run.py", line 785, in read_movie_header
    nz, ny, nx = eer.read_eer_shape(abs_path)
  File "/cryosparc/cryosparc_worker/cryosparc_compute/blobio/eer.py", line 15, in read_eer_shape
    ioengine.eer_get_inherent_size(fname, xyf)
RuntimeError: Couldn't open specified EER file
[Wed, 09 Jul 2025 01:45:34 GMT] [CPU RAM used: 92 MB] [libtiff error] /cryosparc/data01/FoilHole_31243545_Data_31242638_56_20250630_154907_EER.eer: Not a TIFF or MDI file, bad magic number 21333 (0x5355)
17

I updated to the latest v4.7.1-cuda12 because my system has cuda12.8. The problem was however not resolved.

18

You only need the CUDA12 build if you have Blackwell architecture cards. If using Ada, Ampere, Turing or Pascal the CUDA11 build works perfectly.

Bad magic number would appear to indicate that the file is corrupt. How large is the file? Does it open with, e.g. FIJI (EER is just a TIFF wrapper at the most basic level)? I’m curious whether RELION throws a similar error - can you import and motion correct the specified micrograph movie with RELION?

19

I see. I am using RTX4090 cards, not the Blackwell ones. The OS is Ubuntu 24.04. The files were from a different system. Each file varies from 370 - 470 MB. I have not tried them with FIJI or RELION yet.

20

Hi, I’ve encountered a similar issue before — just wanted to share what worked for me.

In my case, I was able to bypass the initial error by using Relion to import the movies first. There was no error at the import step, but the problem occurred during motion correction, where Relion failed and stopped.

By checking the log, I could identify which specific .eer file caused the issue. After deleting that problematic file, I switched to CryoSPARC and re-imported the movies — the process went through without any issues.

Hope this helps!