Image Optimization processing issues – images requested never cleared

  • Resolved Arnd Zanduino

    (@zanduino)


    1 year, 5 months ago

    I have a site with a lot of images that have collected over the years. I’ve been trying now for over a month to get those images processed using the “Image Optimization” process and have purchased fast-queue units as well to get things processed; but the whole procedure is very, very slow.

    This is not because it takes long for the remote server to process images and create WEBP. The problem is that with each batch of images sent to be processed, several are not processed and remain in the queue. So, even though I have a group of 200 with around 1000 images in a batch, around 150 of them get processed within an hour. The rest don’t seem to get processed. I’ve waited 48 hours and those images are still in queue! So I have to press the “clear up unfinished data” button. That works and lets me send more pictures, but it starts again at the lowest image post id and seems to re-process any images with higher numbers, although they were already successfully processed and retrieved.

    This means that I have to manually control the queues for submitting and pulling and cleaning them up between each 200 image batch. It has taken me most of a month to get close to finishing. Then, suddenly, the Percentage completion which showed 97% dropped back down to 10% and it seems that I have to start again.

    I’ve reset everything using the “destroy optimization data” and started again, but the same issues continue to happen, I don’t get a single full group processed. I have changed the settings to not optimize the original images, as what I would really like is to have WEBP created. The same problems persist.

    The QUIC.cloud shows no error messages, but doesn’t tell me how many or which images are queued for processing. I’ve turned on DEBUGging and couldn’t find any error messages that were relevant in the log.

Viewing 9 replies - 1 through 9 (of 9 total)
  • Thread Starter Arnd Zanduino

    (@zanduino)

    I nuked the database and started again today. The cron settings are turned off for both upload and download.

    1. I press “send optimization request” and get notified that 200 groups (800 images) have been submitted.
      06/14/23 11:54:01.386 [147.135.115.64:44840 1 W3I] Images found: 800
      06/14/23 11:54:01.402 [147.135.115.64:44840 1 W3I] Images after duplicated: 800
      06/14/23 11:54:01.404 [147.135.115.64:44840 1 W3I] Images after invalid: 800
      06/14/23 11:54:01.615 [147.135.115.64:44840 1 W3I] Added raw images [total] 800
    2. After several minutes, I get the following information:

    Images requested: 17 groups (31 images)

    After the QUIC.cloud Image Optimization server finishes optimization, it will notify your site to pull the optimized images. This process is automatic.

    Images notified to pull: 195 groups (551 images)

    Since I have the cron job to pull turned off, this display stays constant, even if I wait an hour or wait overnight.

    a) 800 images – 31 requested – 551 notified to pull = 218
    What has happened to those 218 images? Nothing shows in the debug log.
    b) Why do those 17 groups/ 31 images stay forever?
    c) When I pull the notified images, then press the “cleanup” and resubmit, the processing starts, I assume, at the first requested image. That means it re-processes images that have already be processed

    If I repeat this after “cleanup unprocessed data” then another 200 groups / 800 images are sent and I get

    Images requested: 30 groups (42 images)

    After the QUIC.cloud Image Optimization server finishes optimization, it will notify your site to pull the optimized images. This process is automatic.

    Images notified to pull: 188 groups (518 images)

    Plugin Support qtwrk

    (@qtwrk)

    we are aware there are some glitches in version 5.3 with image optimization and we have also prepared a beta version for fixing it , would you like to give it a try ?

    Thread Starter Arnd Zanduino

    (@zanduino)

    Yes, I’d be willing to give a beta version a try.

    Plugin Support qtwrk

    (@qtwrk)

    okay , please go to Toolbox -> beta test -> click “dev” and then upgrade

    please be aware, since it is still a beta version , so it’s always a good idea to do a full backup first.

    after that , please clean up unfinished data and re-send it from there.

    Thread Starter Arnd Zanduino

    (@zanduino)

    I did a complete site backup before activating the most recent beta, “v5.5-rc5”. I cleaned up and turned off the cron and turned on debugging.

    Unfortunately the error still remains, there are always a few groups/images left on the server for processing (according to the front-end) that never get processed. I’ve checked the debug log for any indications and haven’t seen anything. I’ve logged into the QUIC.cloud to see if there are any messages there, there were none.

    I’ve done a couple of manual runs:

    200 groups, 828 images, after processing 32 groups with 54 images left.
    200 groups, 826 images, after processing 29 groups with 39 images left.
    200 groups, 824 images, after processing 31 groups with 47 images left.
    200 groups, 827 images, after processing 37 groups with 42 images left.

    I neglected to mention that I’d tried changing to another processing node a couple of time in the past, that also made no difference.

    Is there a command I can issue to see how many images are actually in the QUIC.Cloud queue?

    Plugin Support qtwrk

    (@qtwrk)

    please create a ticket by mail to support at litespeedtech.com with reference link to this topic , we will investigate further.

    Thread Starter Arnd Zanduino

    (@zanduino)

    I spent a whole day manually submitting, waiting, clearing unprocessed images, then submitting again. I finally got to 100%. From here on going forward, I’ll only be adding about 10 images a day and have turned on the two cron processes. Once I get another hang I’ll submit a support ticket.

    same exact issue still plaguing, why am I purchasing fast pass and the large request package if this shit don’t work

    Thread Starter Arnd Zanduino

    (@zanduino)

    I had to re-process all my images and the same thing happened to me (again). It took over a week to get through all images, and I had the optimization turned off this time and was just generating the .webp ones. I also purchased additional “fast-queue” slots (again) and that didn’t help.

    But I did look at the database table, “litespeed_img_optming” while processing. This get filled with the list of images to fill and the images that “hang” and never get processed all have an empty “server_info” column even when the other records have that value filled. They can then either get deleted in the table or in the GUI. If they are deleted in the table, they are considered to have been processed and won’t get run through again.

    I did create a ticket, but there has been no progress on the problem.

Viewing 9 replies - 1 through 9 (of 9 total)
  • The topic ‘Image Optimization processing issues – images requested never cleared’ is closed to new replies.