Environment:
AirSane version: v0.4.12 (commit 129cc3b)
Platform: Raspberry Pi 3, aarch64 (Debian-based)
SANE backend: hpaio (HPLIP), network connection
Scanner: HP LaserJet 100 color MFP M175nw, network-connected (device URI: hpaio:/net/hp_laserjet_100_colormfp_m175nw?ip=10.78.70.30&queue=false)
Client: macOS Image Capture (eSCL)
Description:
When scanning a multi-page document via the ADF through AirSane's eSCL interface, the scan freezes/hangs after 1-2 pages, requiring the job to be manually cancelled from the client. Flatbed (single-page) scanning works correctly through AirSane with no issues.
Steps to reproduce:
Load 3+ pages in the ADF of the HP M175nw
Scan from macOS Image Capture using the AirSane-published eSCL scanner
Observe: 1st (sometimes 2nd) page scans successfully, then the job hangs indefinitely with no error until manually cancelled
Confirming this is AirSane-specific (not the scanner/backend):
Running scanimage -d "hpaio:/net/hp_laserjet_100_colormfp_m175nw?ip=10.78.70.30&queue=false" --batch=/tmp/page%d.pnm --source=ADF directly on the server successfully scans all pages (tested with 15 pages) without any hang
Using scanservjs with the same raw hpaio device (bypassing AirSane entirely) also successfully completes the full ADF batch (15 pages)
Only the AirSane eSCL path (used by Image Capture) exhibits the freeze
Relevant log output during device enumeration (airsaned --debug=true):
io/hpmud/model.c 534: no laserjet_100_colormfp_m175nw attributes found in /usr/share/hplip/data/models/models.dat
io/hpmud/model.c 545: no laserjet_100_colormfp_m175nw attributes found in /usr/share/hplip/data/models/unreleased/unreleased.dat
common/utils.c 212: unable to load library libm.so: /lib/aarch64-linux-gnu/libm.so: invalid ELF header
bb_soapht.c 295: unknowned element=18122912
The bb_soapht.c: unknowned element warning in particular looks like it could be related to a parsing issue in the ADF "next page" status handling, though I'm not certain this is the root cause.
Happy to provide more logs (e.g. SANE_DEBUG_HPAIO=255 output during an actual multi-page hang) if useful.
Environment:
AirSane version: v0.4.12 (commit 129cc3b)
Platform: Raspberry Pi 3, aarch64 (Debian-based)
SANE backend: hpaio (HPLIP), network connection
Scanner: HP LaserJet 100 color MFP M175nw, network-connected (device URI: hpaio:/net/hp_laserjet_100_colormfp_m175nw?ip=10.78.70.30&queue=false)
Client: macOS Image Capture (eSCL)
Description:
When scanning a multi-page document via the ADF through AirSane's eSCL interface, the scan freezes/hangs after 1-2 pages, requiring the job to be manually cancelled from the client. Flatbed (single-page) scanning works correctly through AirSane with no issues.
Steps to reproduce:
Load 3+ pages in the ADF of the HP M175nw
Scan from macOS Image Capture using the AirSane-published eSCL scanner
Observe: 1st (sometimes 2nd) page scans successfully, then the job hangs indefinitely with no error until manually cancelled
Confirming this is AirSane-specific (not the scanner/backend):
Running scanimage -d "hpaio:/net/hp_laserjet_100_colormfp_m175nw?ip=10.78.70.30&queue=false" --batch=/tmp/page%d.pnm --source=ADF directly on the server successfully scans all pages (tested with 15 pages) without any hang
Using scanservjs with the same raw hpaio device (bypassing AirSane entirely) also successfully completes the full ADF batch (15 pages)
Only the AirSane eSCL path (used by Image Capture) exhibits the freeze
Relevant log output during device enumeration (airsaned --debug=true):
io/hpmud/model.c 534: no laserjet_100_colormfp_m175nw attributes found in /usr/share/hplip/data/models/models.dat
io/hpmud/model.c 545: no laserjet_100_colormfp_m175nw attributes found in /usr/share/hplip/data/models/unreleased/unreleased.dat
common/utils.c 212: unable to load library libm.so: /lib/aarch64-linux-gnu/libm.so: invalid ELF header
bb_soapht.c 295: unknowned element=18122912
The bb_soapht.c: unknowned element warning in particular looks like it could be related to a parsing issue in the ADF "next page" status handling, though I'm not certain this is the root cause.
Happy to provide more logs (e.g. SANE_DEBUG_HPAIO=255 output during an actual multi-page hang) if useful.