Serialization error

Hello!

I did the object detection tutorial (others also) based on this site: https://awesomeopensource.com/project/dusty-nv/jetson-inference#code-examples

It worked perfectly, but now, they don’t work at all. I didn’t change any code, the only difference is I did some recommended system update.

Please help me, how solve the problem, how to get the codes work again. Thank you!

detectNet – loading detection network model from:
– model networks/SSD-Mobilenet-v2/ssd_mobilenet_v2_coco.uff
– input_blob ‘Input’
– output_blob ‘NMS’
– output_count ‘NMS_1’
– class_labels networks/SSD-Mobilenet-v2/ssd_coco_labels.txt
– threshold 0.500000
– batch_size 1

[TRT] TensorRT version 7.1.0
[TRT] loading NVIDIA plugins…
[TRT] Registered plugin creator - ::GridAnchor_TRT version 1
[TRT] Registered plugin creator - ::NMS_TRT version 1
[TRT] Registered plugin creator - ::Reorg_TRT version 1
[TRT] Registered plugin creator - ::Region_TRT version 1
[TRT] Registered plugin creator - ::Clip_TRT version 1
[TRT] Registered plugin creator - ::LReLU_TRT version 1
[TRT] Registered plugin creator - ::PriorBox_TRT version 1
[TRT] Registered plugin creator - ::Normalize_TRT version 1
[TRT] Registered plugin creator - ::RPROI_TRT version 1
[TRT] Registered plugin creator - ::BatchedNMS_TRT version 1
[TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1
[TRT] Registered plugin creator - ::CropAndResize version 1
[TRT] Registered plugin creator - ::DetectionLayer_TRT version 1
[TRT] Registered plugin creator - ::Proposal version 1
[TRT] Registered plugin creator - ::ProposalLayer_TRT version 1
[TRT] Registered plugin creator - ::PyramidROIAlign_TRT version 1
[TRT] Registered plugin creator - ::ResizeNearest_TRT version 1
[TRT] Registered plugin creator - ::Split version 1
[TRT] Registered plugin creator - ::SpecialSlice_TRT version 1
[TRT] Registered plugin creator - ::InstanceNormalization_TRT version 1
[TRT] detected model format - UFF (extension ‘.uff’)
[TRT] desired precision specified for GPU: FASTEST
[TRT] requested fasted precision for device GPU without providing valid calibrator, disabling INT8
[TRT] native precisions detected for GPU: FP32, FP16
[TRT] selecting fastest native precision for GPU: FP16
[TRT] attempting to open engine cache file /usr/local/bin/networks/SSD-Mobilenet-v2/ssd_mobilenet_v2_coco.uff.1.1.7100.GPU.FP16.engine
[TRT] loading network plan from engine cache… /usr/local/bin/networks/SSD-Mobilenet-v2/ssd_mobilenet_v2_coco.uff.1.1.7100.GPU.FP16.engine
[TRT] device GPU, loaded /usr/local/bin/networks/SSD-Mobilenet-v2/ssd_mobilenet_v2_coco.uff
[TRT] coreReadArchive.cpp (38) - Serialization Error in verifyHeader: 0 (Version tag does not match)
[TRT] INVALID_STATE: std::exception
[TRT] INVALID_CONFIG: Deserialize the cuda engine failed.
[TRT] device GPU, failed to create CUDA engine
[TRT] failed to create TensorRT engine for /usr/local/bin/networks/SSD-Mobilenet-v2/ssd_mobilenet_v2_coco.uff, device GPU
[TRT] detectNet – failed to initialize.
detectnet-console: failed to initialize detectNet
Press ENTER to continue…

Hi,

It looks like you are using TensorRT 7.1.0 which should come from JetPack4.4 DP.
It’s known that the version do have some issue in the serializer and the corresponding fix is available in the TensorRT 7.1.3 (JetPack4.4 GA).

Would you mind to upgrade your system into JetPak4.4 GA first:

Thanks.

Hi,

Thank you for your response! I flashed my TX2 with JetPack4.2. Is there any way to upgrade my JetPack4.2 to JetPack4.4 GA without flash my Jetson again? I don’t want to lose my settings and start the configuring from zero.

If there is, please let me know. Thank you!

Hi,

Sorry that you will need to reflash the system for compatibility.

This is because that the GPU driver is updated across different JetPack version.
But Jetson’s GPU driver is integrated into OS which cannot be upgraded separately.

Thanks.

Hi,

Thank you for your response!
My Jetson worked fine, after this update it doesn’t run some of my codes. What should I do in settings to avoid the reflashing in the future? Do not allow any update?

Thank you!

Each upgrade tends to also upgrade the CUDA release version. Some of your software may simply need to be recompiled against the new CUDA release version.

The older (but still somewhat recent) releases did not support OTA update, but the version you just flashed should allow such an upgrade for future releases.

If in doubt back up the system via clone or rsync. The future system cannot directly use the old clone/rsync, but you can use it to repopulate non-o/s content, e.g., “/home”.

1 Like