Support

The log excerpt you pasted does not contain enough information to diagnose the problem. Please provide the complete log file.

Go to Akeeba Backup, Manage Backups.

Find the records of the failed backup and click the View Log button in its right-hand column.

Click the Download Log File button towards the top of the page, next to the log file selector drop-down.

This downloads a .txt file. Put that file inside a ZIP file.

• If the ZIP file is under 10 MiB: attach it to your next reply
• If the ZIP file is over 10 MiB: upload it to Dropbox, then paste me a public share link in your next reply

The log file tells me what is going on during backup – including the upload stage, if it's actually reached that step. Let me look at that file so I can give you a much better suggestion about what to do next.

Regarding table exclusions: the regular Database Table Exclusion filter does not support wildcards. If you want to exclude all tables matching a pattern — for example, all tables whose name starts with sv_bookpro3 — you should use the RegEx Database Table Exclusion filter instead. The corresponding regular expression would be /^#__sv_bookpro3/.

All filters in Akeeba Backup — including the RegEx Database Table Filters — are part of the backup profile. There is no separate "linking" step.

You simply need to make sure the correct profile is selected on the Akeeba Backup main page (the control panel) before navigating to any filter page. Whichever profile is active there is the one your filters apply to. So the workflow is:

1. Go to the Akeeba Backup main page.
2. Select the profile you want to configure from the profile selector.
3. Navigate to the RegEx Database Table Filters page. The regex you enter there will be saved as part of that profile.

This has been how backup profiles work since they were introduced — a profile consists of its configuration options and its filters together.

Regarding the Files and Directories Exclusion page: the meaning of each column and the icons in it are explained in our documentation. The same applies: select the correct profile on the main page first, then configure your directory exclusions.

Regarding notification emails: our site has sent them from [email protected] to the email address on your account. From that point, delivery is in the hands of your email provider and mail client, which is outside our control.

The steps below are exact. Please follow them in order.

Step 1 — Select the correct profile

Go to Components > Akeeba Backup. This takes you to the main Akeeba Backup page.

At the top of that page there is a profile selector drop-down. Click it and select the profile you use for your backups ("Profil EXCLUE 2 tables et media/k2 + images" or whichever one you run the backup with). Do not navigate away from this page yet — just select the profile.

Step 2 — Open the RegEx Database Table Filters

Still on the Akeeba Backup main page, click on RegEx Database Table Filters in the left-hand sidebar. Because you selected the profile in Step 1, this page is now showing the filters for that profile.

Step 3 — Add the exclusion

Find the field labelled Exclude tables (not "exclude rows" or any other option — specifically "exclude tables"). Click the Add button next to it and enter:

/^#__sv_bookpro3/

Then click Save.

Step 4 — Confirm

Run the backup again. The tables whose names start with sv_bookpro3 will now be skipped entirely.

Regarding the Files and Directories Exclusion: our documentation has an index. The section you need is titled Files and Directories Exclusion — which you can find directly at https://www.akeeba.com/documentation/akeeba-backup-joomla/exclude-data-from-backup.html#files-and-directories-exclusion. The same rule applies: select the correct profile on the main page first.

We link to the documentation rather than reproducing it here for three reasons. First, copying it into every ticket would make replies very long and hard to follow. Second, it is written by the same people who answer support tickets — we can only explain the same thing in so many ways. Third, and most importantly, the documentation is kept up to date with every release; a ticket reply is not. Anyone reading this ticket in the future — including yourself — will find current, accurate information by following the link, rather than instructions that may no longer reflect how the software works.

Regarding the complete log file from the original stalled backup: we have asked for it twice now and have not received it. We cannot diagnose why the backup is stalling without it. If we do not receive the complete log file in your next reply, we will have no choice but to close this ticket as unsolvable — not because we do not want to help, but because we simply cannot help without the information needed to troubleshoot.

Looking at the log file you attached, there are two things to address.

The backup is not erroring out. What happened is that the browser stopped receiving responses from the server — the backup engine was still running, but the connection from the progress page to the server went silent. The most common cause is the device going to sleep, or the browser suspending a background tab. The backup progress page itself displays a notice about this at the top of the page. Please keep your device awake and keep the backup progress page visible and active in the foreground for the entire duration of the backup. If your device goes to sleep mid-backup, the progress page will never know the backup finished and the backup will appear to hang.

The log also shows the backup is processing the file images-preprod-PH.zip, which is nearly 7 GiB. Are you certain you want this file included in your backup? If it is a pre-production archive that you do not need to restore from, you can exclude it using the Files and Directories Exclusion filter — remembering to select the correct profile on the main Akeeba Backup page first. A 7 GiB file will add a very significant amount of time to your backup, making it much more likely that the device will go to sleep before it finishes.

Moira responded to you based on your log file. Your log file says that you are backing up the file images-preprod-PH.zip in your site's root which is 6797876849 bytes long.

Here is the excerpt:

INFO    |20260528 15:38:49|Breaking step _before_ large file: /home/.../.../images-preprod-PH.zip - size: 6797876849
DEBUG   |20260528 15:38:49|Kettenrad :: Domain object returned without errors; propagating
DEBUG   |20260528 15:38:49|Kettenrad :: Break flag status: YES
DEBUG   |20260528 15:38:49|----- Finished operation 2 ------
DEBUG   |20260528 15:38:49|Successful Smart algorithm on Akeeba\Engine\Core\Domain\Pack
DEBUG   |20260528 15:38:49|====== Finished Step number 39 ======
DEBUG   |20260528 15:38:49|Kettenrad :: Setting the break flag between domains
DEBUG   |20260528 15:38:49|*** Engine steps batching: Break flag detected.
DEBUG   |20260528 15:38:49|*** Batching of engine steps finished. I will now return control to the caller.
DEBUG   |20260528 15:38:49|No need to sleep; execution time: 216.03417396545 msec; min. exec. time: 0 msec
DEBUG   |20260528 15:38:49|Saving Kettenrad instance backend
DEBUG   |20260528 15:38:50|Kettenrad :: Attempting to load from database (backend) [backend.id-20260528-153542-918039]
DEBUG   |20260528 15:38:50| -- Loaded stored Akeeba Factory (backend) [backend.id-20260528-153542-918039]
DEBUG   |20260528 15:38:50|====== Starting Step number 40 ======
DEBUG   |20260528 15:38:50|Site root: /home/.../.../
DEBUG   |20260528 15:38:50|Kettenrad :: Ticking the domain object
DEBUG   |20260528 15:38:50|Packing files
DEBUG   |20260528 15:38:50|Truncating backup archive file /home/.../.../.../site_exclus-2tables-mediaK2-images_nmh-preprod-aout.88h.ovh-20260528-153542utc.j01 to 44900825 bytes
DEBUG   |20260528 15:38:50|-- Adding images-preprod-PH.zip to archive (source: /home/.../.../images-preprod-PH.zip)

We understand what you intended to do and that this is what you are saying you did, but it's not what you actually ended up doing as per the log file you gave us. You are either wrong about what you did, or you sent us the wrong log file. In either case, we can only respond to you based on the log file you sent us.

Regarding your browser, kindly note that the JavaScript running on your browser is what tells the backup engine on your server to keep going through the backup steps. If the browser puts the tab to sleep, or your device itself goes to sleep, this code stops running. Therefore, the backup stops going forward. It appears stalled in the log without any errors. This is exactly what we observed. Your backup started at 15:35:42 (GMT) and stopped receiving further instructions to proceed at 15:52:14 (GMT). I observe that this is roughly equivalent to what would've happened if you worked on your computer for a couple of minutes, then stepped away to do something else – the default timeout for putting the computer to sleep on both KDE Plasma 6 and recent versions of GNOME is 15 minutes. That's not a random value; it's actually what the EU prescribes as the maximum default device sleep timeout value to be compliant with its energy savings directive. It was a whole thing a few months ago when KDE Plasma developers scrambled to apply this default so as to make laptops shipping with KDE Plasma compliant with EU regulations, i.e. allow them to legally be sold to businesses in the EU (the primary market of these devices).

To prevent your browser from going to sleep, follow these steps:

1. Open Brave.
2. Click the menu button (three horizontal lines, top-right corner).
3. Click Settings.
4. In the left sidebar, click System.
5. Find the toggle labeled "Sleep inactive tabs after X minutes" (or similar wording about background/inactive tabs).
6. Click the toggle to turn it off (it should change from blue/colored to grey).
7. Close the Settings tab. The change saves automatically — no restart needed.

If you don't see that exact option under System, type brave://settings/performance into the address bar and press Enter — the tab-sleeping setting lives on the Performance page, where you can switch it off there instead.

To prevent your computer from going to sleep please refer to your Operating System's documentation. I would normally give you some sort of instructions, but I see you're running Linux. The exact instructions vary depending on distribution and desktop environment. KDE Plasma 6, which I use, has a very handy feature in its Power Settings tray icon which allows you to prevent your computer from going to sleep and it's accessible with one click. I believe relatively recent versions of GNOME also have something similar. I have not used other DEs in the past several years, I'm afraid, so I can't tell you what they can and cannot do.

I honestly don't know how else to say this without sounding condescending, but I'll try.

This has never been about excluding an image folder.

It's about a massive ZIP file with images in your site's root.

The massive ZIP file is named images-preprod-PH.zip.

The ZIP file is unrelated to and outside your images folder. Excluding the images folder (or any of its subdirectories) will not exclude a file outside that folder.