From 868fc526c8dffffdd4f41bebaf02d72ed6b25cdb Mon Sep 17 00:00:00 2001
From: "stepan.moc" <stepan.moc@actis.cz>
Date: Sun, 15 Feb 2026 20:03:11 +0100
Subject: [PATCH 01/10] fix(git): remove branch rebase bug

---
 ingress_server/setup_venv.sh       | 28 ++++++++++++++++++++++++++++
 ingress_server/tools/manual_run.py | 14 ++++++++++++++
 2 files changed, 42 insertions(+)
 create mode 100644 ingress_server/setup_venv.sh
 create mode 100644 ingress_server/tools/manual_run.py

diff --git a/ingress_server/setup_venv.sh b/ingress_server/setup_venv.sh
new file mode 100644
index 00000000..42d2ddb4
--- /dev/null
+++ b/ingress_server/setup_venv.sh
@@ -0,0 +1,28 @@
+#!/bin/bash
+# This would create a development virtual environment
+# - uses requirements.txt
+# - install endorse itself in development mode.
+set -x
+
+echo "Creating python virtual environment."
+
+SCRIPTPATH="$( cd -- "$(dirname "$0")" >/dev/null 2>&1 || exit ; pwd -P )"
+SRC_ROOT="$SCRIPTPATH"
+
+cd ${SRC_ROOT} || exit
+rm -r venv
+#virtualenv venv
+python3 -m venv --system-site-packages venv
+ls
+
+venv_pip=${SRC_ROOT}/venv/bin/pip
+$venv_pip install wheel
+$venv_pip install --upgrade pip
+#source venv/bin/activate
+$venv_pip install "Flask-SQLAlchemy>=3.0.1"
+#$venv_pip install -r requirements.txt
+
+# TODO simplyfy dependency and submodules
+# attrs somehow was broken after gmsh explicit installation, must force its reinstalation
+
+$venv_pip install -e .
diff --git a/ingress_server/tools/manual_run.py b/ingress_server/tools/manual_run.py
new file mode 100644
index 00000000..fca94ec6
--- /dev/null
+++ b/ingress_server/tools/manual_run.py
@@ -0,0 +1,14 @@
+import sys
+from ingress_server.worker import _process_one, _iter_accepted_files_in_dir
+
+def main():
+    accepted_dir = sys.argv[1]
+    for data_path in _iter_accepted_files_in_dir(accepted_dir):
+        err = _process_one(data_path)
+        if err:
+            print(f"Error processing {data_path}: {err}")
+        else:
+            print(f"Successfully processed {data_path}")
+
+if __name__ == "__main__":
+    main()

From 1840d667c47f8415f3aa8b0e36f2b47f64d2025b Mon Sep 17 00:00:00 2001
From: Jan Brezina <jan.brezina@tul.cz>
Date: Mon, 16 Feb 2026 23:31:52 +0100
Subject: [PATCH 02/10] S3 CESNET bandwidth measurements

---
 tests/s3_bandwidth/fs_bandwidth.sh      |   6 +
 tests/s3_bandwidth/results.md           |  70 ++++++++++
 tests/s3_bandwidth/setup.sh             |  21 +++
 tests/s3_bandwidth/test_s3_bandwidth.py | 166 ++++++++++++++++++++++++
 4 files changed, 263 insertions(+)
 create mode 100644 tests/s3_bandwidth/fs_bandwidth.sh
 create mode 100644 tests/s3_bandwidth/results.md
 create mode 100644 tests/s3_bandwidth/setup.sh
 create mode 100644 tests/s3_bandwidth/test_s3_bandwidth.py

diff --git a/tests/s3_bandwidth/fs_bandwidth.sh b/tests/s3_bandwidth/fs_bandwidth.sh
new file mode 100644
index 00000000..0882a67a
--- /dev/null
+++ b/tests/s3_bandwidth/fs_bandwidth.sh
@@ -0,0 +1,6 @@
+FILE="$1/test_file"
+dd if=/dev/zero of="$FILE" bs=64M count=16 oflag=direct status=progress
+
+dd if="$FILE" of=/dev/null bs=64M iflag=direct status=progress
+
+rm -f "$FILE"
diff --git a/tests/s3_bandwidth/results.md b/tests/s3_bandwidth/results.md
new file mode 100644
index 00000000..33313b64
--- /dev/null
+++ b/tests/s3_bandwidth/results.md
@@ -0,0 +1,70 @@
+# Theoretical and practical bandwidth for CESNET
+
+## CESNET Ceph CL4 info
+
+CL4 (Brno)
+
+Ceph-based object storage
+
+57 storage servers
+
+24 HDDs per server
+
+Total raw capacity: 26.607 PB
+
+Access layer: 5 front-end servers + 9 application servers
+
+Theoretical max bandwidth per client 1-2 Gbps
+
+
+## Brno - Charon network bandwidth
+
+Skirit (Brno) - Charon (Liberec) Bandwidth
+(BOOKWORM)jan_brezina@skirit:~/lbc/workspace/zarr_test$ bash fs_bandwidth.sh .
+
+write to charon, 16 x 64MB
+1073741824 bytes (1,1 GB, 1,0 GiB) copied, 18,6463 s, 57,6 MB/s
+
+read from charon 16x64MB
+1073741824 bytes (1,1 GB, 1,0 GiB) copied, 15,4808 s, 69,4 MB/s
+
+## S3 access from Charon, Liberec
+
+Charon - CESNET (Liberec, 1GBit
+
+Endpoint: https://s3.cl4.du.cesnet.cz
+Bucket:   test-zarr-storage
+Key:      bwtest/38c648e951bf461bad38f473ba22758e.bin
+Size:     1073741824 bytes (1 GiB)
+
+UPLOAD:   26.474 s  -> 38.7 MiB/s (0.32 Gbit/s)
+DOWNLOAD: 7.460 s  -> 137.3 MiB/s (1.15 Gbit/s)
+
+UPLOAD:   25.947 s  -> 39.5 MiB/s (0.33 Gbit/s)
+DOWNLOAD: 7.133 s  -> 143.6 MiB/s (1.20 Gbit/s)
+
+
+## S3 access from Skirit, Brno
+
+Endpoint: https://s3.cl4.du.cesnet.cz
+Bucket:   test-zarr-storage
+Key:      bwtest/d20d78d0be4a4520b1e4e9a94cde1cff.bin
+Size:     1073741824 bytes (1 GiB)
+
+UPLOAD:   23.865 s  -> 42.9 MiB/s (0.36 Gbit/s)
+DOWNLOAD: 5.664 s  -> 180.8 MiB/s (1.52 Gbit/s)
+
+UPLOAD:   24.012 s  -> 42.6 MiB/s (0.36 Gbit/s)
+DOWNLOAD: 5.478 s  -> 186.9 MiB/s (1.57 Gbit/s)
+
+## Conclusions
+
+- CESNET [documents](https://docs.du.cesnet.cz/en/docs/object-storage-s3/s5cmd?utm_source=chatgpt.com) peak bandwidth for 1-2Gps netwok connections using `s5cmd` util
+- The network itself is capable of such bandwidth (Charon - Brno)
+- For read we are close to the limit, which in fact is very close to the limit of mechanical HDDs.
+- For write there is probably room for improvement up to 4-5 times.
+- Storage has 5 frontends so it probably can handle multiple parallel writes. 
+
+## TODO:
+- Measure serial zarr and zarr-fuse R/W bandwidth.
+- Measure real parallel read, parallel write bandwidth.
diff --git a/tests/s3_bandwidth/setup.sh b/tests/s3_bandwidth/setup.sh
new file mode 100644
index 00000000..41b0c4df
--- /dev/null
+++ b/tests/s3_bandwidth/setup.sh
@@ -0,0 +1,21 @@
+#!/usr/bin/env bash
+set -xeuo pipefail
+
+module load python
+
+VENV_DIR="venv"
+
+# Create venv if missing
+if [[ ! -d "$VENV_DIR" ]]; then
+  python3 -m venv "$VENV_DIR"
+fi
+
+# Activate venv
+# shellcheck disable=SC1090
+#source "$VENV_DIR/bin/activate"
+
+PY="$VENV_DIR/bin/python3"
+# Minimal deps for the S3 test script
+$PY -m pip install --upgrade pip >/dev/null
+$PY -m pip install boto3
+
diff --git a/tests/s3_bandwidth/test_s3_bandwidth.py b/tests/s3_bandwidth/test_s3_bandwidth.py
new file mode 100644
index 00000000..552b3f58
--- /dev/null
+++ b/tests/s3_bandwidth/test_s3_bandwidth.py
@@ -0,0 +1,166 @@
+#!/usr/bin/env python3
+import json
+import math
+import sys
+import time
+import uuid
+
+import boto3
+from botocore.config import Config
+
+# === EDIT THESE TWO ===
+S3_URL = "https://s3.cl4.du.cesnet.cz"   # e.g. "https://s3.cld.cesnet.cz"
+BUCKET = "test-zarr-storage"
+# ======================
+
+GiB = 1024 ** 3
+MiB = 1024 ** 2
+
+SIZE_BYTES = 1 * GiB
+PART_SIZE = 64 * MiB   # must be >= 5 MiB for multipart
+DL_CHUNK = 8 * MiB
+
+
+def die(msg: str, code: int = 2) -> None:
+    print(f"ERROR: {msg}", file=sys.stderr)
+    raise SystemExit(code)
+
+
+def human_speed(bytes_per_sec: float) -> str:
+    mib_s = bytes_per_sec / MiB
+    gbit_s = (bytes_per_sec * 8) / 1e9
+    return f"{mib_s:.1f} MiB/s ({gbit_s:.2f} Gbit/s)"
+
+
+def make_client(creds: dict):
+    ak = creds.get("access_key")
+    sk = creds.get("secret_key")
+    st = creds.get("session_token")
+
+    if not ak or not sk:
+        die('Secrets JSON must include "access_key" and "secret_key".')
+
+    # CESNET / S3-compatible compatibility knobs:
+    # - region_name is used for SigV4 signing only
+    # - checksum settings avoid AWS-style streaming checksums/trailers that some endpoints reject
+    cfg = Config(
+        region_name="us-east-1",
+        s3={"addressing_style": "path"},
+        retries={"max_attempts": 5, "mode": "standard"},
+        request_checksum_calculation="when_required",
+        response_checksum_validation="when_required",
+    )
+
+    return boto3.client(
+        "s3",
+        endpoint_url=S3_URL,
+        aws_access_key_id=ak,
+        aws_secret_access_key=sk,
+        aws_session_token=st,
+        config=cfg,
+    )
+
+
+def multipart_upload(s3, bucket: str, key: str) -> float:
+    resp = s3.create_multipart_upload(Bucket=bucket, Key=key)
+    upload_id = resp["UploadId"]
+
+    buf = b"\0" * PART_SIZE
+    num_parts = math.ceil(SIZE_BYTES / PART_SIZE)
+
+    parts = []
+    sent = 0
+    start = time.perf_counter()
+
+    try:
+        for part_number in range(1, num_parts + 1):
+            remaining = SIZE_BYTES - sent
+            this_size = PART_SIZE if remaining >= PART_SIZE else remaining
+            body = buf if this_size == PART_SIZE else (b"\0" * this_size)
+
+            up = s3.upload_part(
+                Bucket=bucket,
+                Key=key,
+                PartNumber=part_number,
+                UploadId=upload_id,
+                Body=body,
+                ContentLength=this_size,  # important for some S3-compatible servers (avoids MissingContentLength)
+            )
+            parts.append({"PartNumber": part_number, "ETag": up["ETag"]})
+            sent += this_size
+
+        s3.complete_multipart_upload(
+            Bucket=bucket,
+            Key=key,
+            UploadId=upload_id,
+            MultipartUpload={"Parts": parts},
+        )
+    except Exception:
+        try:
+            s3.abort_multipart_upload(Bucket=bucket, Key=key, UploadId=upload_id)
+        except Exception:
+            pass
+        raise
+
+    return time.perf_counter() - start
+
+
+def streaming_download(s3, bucket: str, key: str) -> float:
+    start = time.perf_counter()
+    obj = s3.get_object(Bucket=bucket, Key=key)
+    body = obj["Body"]
+
+    while True:
+        chunk = body.read(DL_CHUNK)
+        if not chunk:
+            break
+
+    body.close()
+    return time.perf_counter() - start
+
+
+def main():
+    if len(sys.argv) != 2:
+        die("Usage: python3 test_s3_bandwidth.py /path/to/secrets.json")
+
+    # Read secrets from JSON file (single positional param = file path)
+    with open(sys.argv[1], "r", encoding="utf-8") as f:
+        raw = f.read()
+
+    creds = json.loads(raw)
+    if not isinstance(creds, dict):
+        die("Secrets JSON must be an object/dict.")
+
+    if not S3_URL.startswith("http"):
+        die("Set S3_URL in the script (must start with http/https).")
+    if not BUCKET:
+        die("Set BUCKET in the script.")
+
+    s3 = make_client(creds)
+    key = f"bwtest/{uuid.uuid4().hex}.bin"
+
+    print(f"Endpoint: {S3_URL}")
+    print(f"Bucket:   {BUCKET}")
+    print(f"Key:      {key}")
+    print(f"Size:     {SIZE_BYTES} bytes (1 GiB)")
+    print()
+
+    # Upload
+    t_up = multipart_upload(s3, BUCKET, key)
+    up_bps = SIZE_BYTES / t_up
+    print(f"UPLOAD:   {t_up:.3f} s  -> {human_speed(up_bps)}")
+
+    # Download
+    t_dn = streaming_download(s3, BUCKET, key)
+    dn_bps = SIZE_BYTES / t_dn
+    print(f"DOWNLOAD: {t_dn:.3f} s  -> {human_speed(dn_bps)}")
+
+    # Cleanup
+    try:
+        s3.delete_object(Bucket=BUCKET, Key=key)
+    except Exception:
+        pass
+
+
+if __name__ == "__main__":
+    main()

From ffe7c3d192bf056ac12af51ea77035a16a6dcd76 Mon Sep 17 00:00:00 2001
From: Jan Brezina <jan.brezina@tul.cz>
Date: Tue, 17 Feb 2026 01:29:50 +0100
Subject: [PATCH 03/10] Revert dasboard deployments

---
 .github/workflows/dashboard-pull-request.yaml | 19 +++++++++++++++++++
 .github/workflows/dashboard-push-main.yaml    | 19 +++++++++++++++++++
 2 files changed, 38 insertions(+)
 create mode 100644 .github/workflows/dashboard-pull-request.yaml
 create mode 100644 .github/workflows/dashboard-push-main.yaml

diff --git a/.github/workflows/dashboard-pull-request.yaml b/.github/workflows/dashboard-pull-request.yaml
new file mode 100644
index 00000000..8f0c820f
--- /dev/null
+++ b/.github/workflows/dashboard-pull-request.yaml
@@ -0,0 +1,19 @@
+name: Dashboard - pull request
+on:
+  pull_request:
+    branches:
+      - "**"
+    paths:
+      - app/databuk/dashboard/**
+  workflow_dispatch:
+jobs:
+  build:
+    name: Build and lint dashboard
+    uses: ./.github/workflows/dashboard-reusable-workflow.yaml
+    with:
+      tag: generate
+      deploy: true
+      namespace: zarr-fuse-dashboard-development
+      release-name: dashboard-development
+      s3-bucket-name: hlavo-release
+    secrets: inherit
diff --git a/.github/workflows/dashboard-push-main.yaml b/.github/workflows/dashboard-push-main.yaml
new file mode 100644
index 00000000..84b16c8b
--- /dev/null
+++ b/.github/workflows/dashboard-push-main.yaml
@@ -0,0 +1,19 @@
+name: Dashboard - push main
+on:
+  push:
+    branches:
+      - dashboard-main
+    paths:
+      - app/databuk/dashboard/**
+  workflow_dispatch:
+jobs:
+  build-and-deploy:
+    name: Build and lint dashboard
+    uses: ./.github/workflows/dashboard-reusable-workflow.yaml
+    with:
+      tag: generate
+      deploy: true
+      namespace: zarr-fuse-dashboard
+      release-name: dashboard
+      s3-bucket-name: hlavo-release
+    secrets: inherit

From 0b055f3a16d843b9432a88b277cadc577d9c28df Mon Sep 17 00:00:00 2001
From: Jan Brezina <jan.brezina@tul.cz>
Date: Tue, 17 Feb 2026 01:33:42 +0100
Subject: [PATCH 04/10] Revert databuk/input/__init__.py

---
 app/databuk/inputs/__init__.py | 11 +++++++++++
 1 file changed, 11 insertions(+)
 create mode 100644 app/databuk/inputs/__init__.py

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
new file mode 100644
index 00000000..e122ab74
--- /dev/null
+++ b/app/databuk/inputs/__init__.py
@@ -0,0 +1,11 @@
+import pathlib
+
+# common output driectory
+__script_dir__ = pathlib.Path(__file__).parent
+work_dir = __script_dir__.parent / "workdir"
+work_dir.mkdir(parents=True, exist_ok=True)
+
+# Following is public
+input_dir = __script_dir__.parent / "inputs"
+schema_bukov_yaml = input_dir / "schema_bukov.yaml"
+bukov_locations_csv = input_dir / "bukov_locations.csv"

From b3b6a2fd13127551bfd54949ea457ae5341be592 Mon Sep 17 00:00:00 2001
From: "stepan.moc" <stepan.moc@actis.cz>
Date: Sun, 22 Feb 2026 16:00:10 +0100
Subject: [PATCH 05/10] fix(ci,oci): resolve threads

---
 .../dashboard-reusable-workflow.yaml          | 16 ++++-----
 .../ingress-server-pull-request.yaml          |  3 +-
 .../workflows/ingress-server-push-main.yaml   |  1 +
 .../workflows/ingress-server-push-tag.yaml    |  3 +-
 .../ingress-server-reusable-workflow.yaml     | 34 ++++++-------------
 ingress_server/oci/Containerfile              |  4 +--
 ingress_server/oci/install_extras.sh          |  4 +++
 7 files changed, 30 insertions(+), 35 deletions(-)

diff --git a/.github/workflows/dashboard-reusable-workflow.yaml b/.github/workflows/dashboard-reusable-workflow.yaml
index ab756e9f..d0ae79c3 100644
--- a/.github/workflows/dashboard-reusable-workflow.yaml
+++ b/.github/workflows/dashboard-reusable-workflow.yaml
@@ -36,7 +36,7 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Get version tag
         id: get_version_tag
         run: |
@@ -61,7 +61,7 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Containerize
         uses: ./.github/actions/containerize
         with:
@@ -87,7 +87,7 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Install dependencies
         run: npm ci
       - name: Build frontend
@@ -95,7 +95,7 @@ jobs:
           VITE_API_URL: "https://zarr-fuse-${{ inputs.release-name }}.dyn.cloud.e-infra.cz"
         run: npm run build
       - name: Upload artifact
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: dashboard-frontend
           path: app/databuk/dashboard/dist
@@ -117,9 +117,9 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Download artifact
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v7
         with:
           name: dashboard-frontend
           path: app/databuk/dashboard/dist
@@ -146,7 +146,7 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Lint helm chart
         uses: ./.github/actions/helm-lint
         with:
@@ -172,7 +172,7 @@ jobs:
       cancel-in-progress: false
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
       - name: Decode kubeconfig
         run: echo "${{ secrets.KUBECONFIG }}" | base64 -d > kubeconfig.yaml && chmod 600 kubeconfig.yaml
       - name: Ensure that the namespace exists
diff --git a/.github/workflows/ingress-server-pull-request.yaml b/.github/workflows/ingress-server-pull-request.yaml
index 169c59d8..ed9e5890 100644
--- a/.github/workflows/ingress-server-pull-request.yaml
+++ b/.github/workflows/ingress-server-pull-request.yaml
@@ -10,10 +10,11 @@ jobs:
     name: Build and deploy ingress server
     uses: ./.github/workflows/ingress-server-reusable-workflow.yaml
     concurrency:
-      group: "${{ github.workflow }}-${{ github.ref_name }}-build-and-deploy"
+      group: "${{ github.workflow }}-${{ github.head_ref }}-build-and-deploy"
       cancel-in-progress: false
     with:
       deploy: false
       push-container: false
       tag: generate
+      zarr-fuse-ref: ${{ github.head_ref }}
     secrets: inherit
diff --git a/.github/workflows/ingress-server-push-main.yaml b/.github/workflows/ingress-server-push-main.yaml
index 0ac241b9..0dc6ed41 100644
--- a/.github/workflows/ingress-server-push-main.yaml
+++ b/.github/workflows/ingress-server-push-main.yaml
@@ -18,4 +18,5 @@ jobs:
       namespace: ingress-server-latest
       release-name: ingress-server-latest
       s3-store-url: s3://app-databuk-test-service/bukov-main.zarr
+      zarr-fuse-ref: ${{ github.ref_name }}
     secrets: inherit
diff --git a/.github/workflows/ingress-server-push-tag.yaml b/.github/workflows/ingress-server-push-tag.yaml
index 7429ff16..3c3b69ae 100644
--- a/.github/workflows/ingress-server-push-tag.yaml
+++ b/.github/workflows/ingress-server-push-tag.yaml
@@ -14,7 +14,7 @@ jobs:
       cancel-in-progress: true
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
 
       - name: Get version from tag
         id: get-version
@@ -36,4 +36,5 @@ jobs:
       namespace: ingress-server
       release-name: ingress-server
       s3-store-url: s3://app-databuk-release-service/bukov.zarr
+      zarr-fuse-ref: ${{ github.ref_name }}
     secrets: inherit
diff --git a/.github/workflows/ingress-server-reusable-workflow.yaml b/.github/workflows/ingress-server-reusable-workflow.yaml
index aa6b4955..343ad235 100644
--- a/.github/workflows/ingress-server-reusable-workflow.yaml
+++ b/.github/workflows/ingress-server-reusable-workflow.yaml
@@ -53,7 +53,7 @@ on:
         description: "Ref/branch/tag repozitáře GeoMop/zarr_fuse for external use"
         required: false
         type: string
-        default: ""
+        default: "main"
 jobs:
   get-version-tag:
     name: Get version tag
@@ -63,6 +63,9 @@ jobs:
     steps:
       - name: Checkout zarr_fuse
         uses: actions/checkout@v4
+        with:
+          repository: GeoMop/zarr_fuse
+          ref: ${{ inputs.zarr-fuse-ref }}
 
       - name: Get version tag
         id: get_version_tag
@@ -78,7 +81,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout caller repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
 
       - name: Copy configuration files
         run: |
@@ -95,7 +98,7 @@ jobs:
 
           cp -r "$SRC_DIR"/. "$DEST_DIR"/
       - name: Upload configuration files as artifact
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: ingress-server-configuration
           path: ci-inputs/
@@ -111,12 +114,7 @@ jobs:
       options: --privileged
     steps:
       - name: Checkout zarr_fuse
-        if: ${{ github.repository == 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
-
-      - name: Checkout zarr_fuse for external use
-        if: ${{ github.repository != 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           repository: GeoMop/zarr_fuse
           ref: ${{ inputs.zarr-fuse-ref }}
@@ -127,7 +125,7 @@ jobs:
 
       - name: Download configuration files artifact
         if: ${{ inputs.configuration-dir-path != '' }}
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v7
         with:
           name: ingress-server-configuration
           path: ingress_server/inputs/
@@ -154,12 +152,7 @@ jobs:
       - containerize
     steps:
       - name: Checkout zarr_fuse
-        if: ${{ github.repository == 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
-
-      - name: Checkout zarr_fuse for external use
-        if: ${{ github.repository != 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           repository: GeoMop/zarr_fuse
           ref: ${{ inputs.zarr-fuse-ref }}
@@ -170,7 +163,7 @@ jobs:
           working-directory: ingress_server/charts/ingress-server
           values: --values values/minimal-required-values.yaml
 
-  eploy:
+  deploy:
     name: Deploy ingress server to e-infra rancher
     runs-on: ubuntu-latest
     if: ${{ inputs.deploy }}
@@ -186,12 +179,7 @@ jobs:
         working-directory: ingress_server/charts
     steps:
       - name: Checkout zarr_fuse
-        if: ${{ github.repository == 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
-
-      - name: Checkout zarr_fuse for external use
-        if: ${{ github.repository != 'GeoMop/zarr_fuse' }}
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           repository: GeoMop/zarr_fuse
           ref: ${{ inputs.zarr-fuse-ref }}
diff --git a/ingress_server/oci/Containerfile b/ingress_server/oci/Containerfile
index f9104f1f..db98b05f 100644
--- a/ingress_server/oci/Containerfile
+++ b/ingress_server/oci/Containerfile
@@ -2,7 +2,7 @@ FROM docker.io/library/python:3.11-slim-bullseye
 
 ARG APP_VERSION="devel"
 
-ARG zarr_fuse_version="main"
+ARG zarr_fuse_branch="main"
 ARG git_version="1:2.30.2-1+deb11u5"
 ARG waitress_version="3.0.2"
 
@@ -25,7 +25,7 @@ RUN groupadd -r ingress && useradd -r -g ingress -u 11233 -m -d /ingress-server
     apt-get update && apt-get install -y --no-install-recommends git=${git_version} && \
     apt-get clean && rm -rf /var/lib/apt/lists/*
 
-RUN pip install --no-cache-dir "git+https://github.com/GeoMop/zarr_fuse.git@${zarr_fuse_version}" && \
+RUN pip install --no-cache-dir "git+https://github.com/GeoMop/zarr_fuse.git@${zarr_fuse_branch}" && \
     pip install --no-cache-dir waitress==${waitress_version}
 
 # --- Install ingress server ---
diff --git a/ingress_server/oci/install_extras.sh b/ingress_server/oci/install_extras.sh
index 25b70c70..0b1040c5 100644
--- a/ingress_server/oci/install_extras.sh
+++ b/ingress_server/oci/install_extras.sh
@@ -1,4 +1,8 @@
 #!/bin/sh
+
+# The script is used to install extra dependencies for ingress server extractirs and is used
+# in the OCI (Dockerfile) to improve readability and maintainability of the Dockerfile.
+
 set -eu
 set -x
 

From 1321467b9b02a3e989a68c4986a3f23e23bb8f5f Mon Sep 17 00:00:00 2001
From: "stepan.moc" <stepan.moc@actis.cz>
Date: Sun, 22 Feb 2026 16:36:27 +0100
Subject: [PATCH 06/10] feat(ingress_server): self review

---
 app/databuk/inputs/__init__.py                | 11 ----------
 app/databuk/inputs/paths.py                   | 22 -------------------
 app/databuk/inputs/pyproject.toml             | 11 ----------
 ingress_server/README.MD                      |  6 ++---
 ingress_server/ingress_server/worker.py       | 22 +++++++++----------
 ingress_server/oci/Containerfile              |  6 ++---
 ...s.sh => install_extractor_dependencies.sh} |  0
 7 files changed, 17 insertions(+), 61 deletions(-)
 delete mode 100644 app/databuk/inputs/__init__.py
 delete mode 100644 app/databuk/inputs/paths.py
 delete mode 100644 app/databuk/inputs/pyproject.toml
 rename ingress_server/oci/{install_extras.sh => install_extractor_dependencies.sh} (100%)

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
deleted file mode 100644
index e122ab74..00000000
--- a/app/databuk/inputs/__init__.py
+++ /dev/null
@@ -1,11 +0,0 @@
-import pathlib
-
-# common output driectory
-__script_dir__ = pathlib.Path(__file__).parent
-work_dir = __script_dir__.parent / "workdir"
-work_dir.mkdir(parents=True, exist_ok=True)
-
-# Following is public
-input_dir = __script_dir__.parent / "inputs"
-schema_bukov_yaml = input_dir / "schema_bukov.yaml"
-bukov_locations_csv = input_dir / "bukov_locations.csv"
diff --git a/app/databuk/inputs/paths.py b/app/databuk/inputs/paths.py
deleted file mode 100644
index 4502138e..00000000
--- a/app/databuk/inputs/paths.py
+++ /dev/null
@@ -1,22 +0,0 @@
-import pathlib
-
-
-# common output driectory
-__script_dir__ = pathlib.Path(__file__).parent
-
-# Following is public
-input_dir = __script_dir__
-schema_bukov_yaml = input_dir / "schemas" / "bukov_schema.yaml"
-bukov_locations_csv = input_dir / "bukov_locations.csv"
-
-# test data
-test_dir = __script_dir__ / "test"
-work_dir = test_dir / "workdir"
-work_dir.mkdir(parents=True, exist_ok=True)
-
-measurements_dir = test_dir / "test_measurements"
-_measurements = ['20250915T111522_824a7f3dc0ad.json', '20250915T115149_8b4f1f4535aa.json',
-'20250915T133948_121e738c86ab.json', 'T_123_partial.json']
-old_measurements = [measurements_dir / _measurements[-1]]
-new_measurements = [measurements_dir / _m for _m in _measurements[:-1]]
-
diff --git a/app/databuk/inputs/pyproject.toml b/app/databuk/inputs/pyproject.toml
deleted file mode 100644
index b619444e..00000000
--- a/app/databuk/inputs/pyproject.toml
+++ /dev/null
@@ -1,11 +0,0 @@
-[project]
-name = "extract"
-version = "0.1.0"
-description = "Custom conversion module for ingress server"
-
-[build-system]
-requires = ["setuptools"]
-build-backend = "setuptools.build_meta"
-
-[tool.setuptools]
-packages = ["extract"]
diff --git a/ingress_server/README.MD b/ingress_server/README.MD
index 68440888..c368d296 100644
--- a/ingress_server/README.MD
+++ b/ingress_server/README.MD
@@ -28,8 +28,9 @@ There is **no UI**; interaction happens via HTTP endpoints (multipart uploads).
 ```
 .
 ├─ ingress_server/
-│  ├─ src/
+│  ├─ ingress_server/
 │  │  ├─ main.py                  # Flask backend server
+|  ├─ inputs/
 │  │  ├─ endpoints_config.yaml    # List of API endpoints and their schema paths
 │  │  └─ schemas/                 # Schema files (prod/ci_test)
 │  ├─ docs/
@@ -146,8 +147,7 @@ The container already includes `schemas/` and `endpoints_config.yaml` from the b
 If you want to override them without rebuilding, mount them as read-only volumes:
 
 ```bash
--v $(pwd)/inputs/schemas:/ingress-server/schemas:ro \
--v $(pwd)/inputs/endpoints_config.yaml:/ingress-server/endpoints_config.yaml:ro
+-v $(pwd)/inputs/:/ingress-server/:ro
 ```
 
 ---
diff --git a/ingress_server/ingress_server/worker.py b/ingress_server/ingress_server/worker.py
index 438c1138..ff2f4628 100644
--- a/ingress_server/ingress_server/worker.py
+++ b/ingress_server/ingress_server/worker.py
@@ -36,16 +36,6 @@ def _move_tree_contents(src: Path, dst: Path):
             except OSError:
                 pass
 
-
-def _iter_accepted_files_in_dir(dir: Path):
-    for root, _, files in os.walk(dir):
-        for name in files:
-            if name.endswith(".meta.json"):
-                continue
-            LOG.info("Found accepted file: %s", Path(root) / name)
-            yield Path(root) / name
-
-
 def _iter_accepted_files():
     settings = get_settings()
 
@@ -53,7 +43,17 @@ def _iter_accepted_files():
         LOG.warning("Accepted directory does not exist: %s", settings.accepted_dir)
         yield from ()
         return
-    yield from _iter_accepted_files_in_dir(settings.accepted_dir)
+
+    paths: list[Path] = []
+    for root, _, files in os.walk(settings.accepted_dir):
+        for name in files:
+            if name.endswith(".meta.json"):
+                continue
+            paths.append(Path(root) / name)
+
+    for path in sorted(paths, key=lambda p: p.name):
+        LOG.info("Found accepted file: %s", path)
+        yield path
 
 
 def _load_metadata(data_path: Path) -> tuple[MetadataModel | None, str | None]:
diff --git a/ingress_server/oci/Containerfile b/ingress_server/oci/Containerfile
index db98b05f..071e40bf 100644
--- a/ingress_server/oci/Containerfile
+++ b/ingress_server/oci/Containerfile
@@ -32,12 +32,12 @@ RUN pip install --no-cache-dir "git+https://github.com/GeoMop/zarr_fuse.git@${za
 WORKDIR /ingress-server
 
 COPY --chown=ingress:ingress pyproject.toml ingress_server/ inputs/ ./
-COPY --chown=ingress:ingress oci/install_extras.sh ./scripts/install_extras.sh
+COPY --chown=ingress:ingress oci/install_extractor_dependencies.sh ./scripts/install_extractor_dependencies.sh
 
 RUN pip install --no-cache-dir .
 
-RUN  chmod +x ./scripts/install_extras.sh && \
-    ./scripts/install_extras.sh
+RUN  chmod +x ./scripts/install_extractor_dependencies.sh && \
+    ./scripts/install_extractor_dependencies.sh
 
 ENV CONFIG_DIR_PATH="/ingress-server/inputs"
 
diff --git a/ingress_server/oci/install_extras.sh b/ingress_server/oci/install_extractor_dependencies.sh
similarity index 100%
rename from ingress_server/oci/install_extras.sh
rename to ingress_server/oci/install_extractor_dependencies.sh

From ae3e9617601c270e3b46e41c602605b8b502b60c Mon Sep 17 00:00:00 2001
From: Jan Brezina <jan.brezina@tul.cz>
Date: Tue, 17 Feb 2026 01:33:42 +0100
Subject: [PATCH 07/10] Revert databuk/input/__init__.py

---
 app/databuk/inputs/__init__.py | 11 +++++++++++
 1 file changed, 11 insertions(+)
 create mode 100644 app/databuk/inputs/__init__.py

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
new file mode 100644
index 00000000..e122ab74
--- /dev/null
+++ b/app/databuk/inputs/__init__.py
@@ -0,0 +1,11 @@
+import pathlib
+
+# common output driectory
+__script_dir__ = pathlib.Path(__file__).parent
+work_dir = __script_dir__.parent / "workdir"
+work_dir.mkdir(parents=True, exist_ok=True)
+
+# Following is public
+input_dir = __script_dir__.parent / "inputs"
+schema_bukov_yaml = input_dir / "schema_bukov.yaml"
+bukov_locations_csv = input_dir / "bukov_locations.csv"

From 9fda37bc775f5ceb6650aadc4c1e749186786fb4 Mon Sep 17 00:00:00 2001
From: "stepan.moc" <stepan.moc@actis.cz>
Date: Sun, 22 Feb 2026 16:36:27 +0100
Subject: [PATCH 08/10] feat(ingress_server): self review

---
 app/databuk/inputs/__init__.py | 11 -----------
 1 file changed, 11 deletions(-)
 delete mode 100644 app/databuk/inputs/__init__.py

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
deleted file mode 100644
index e122ab74..00000000
--- a/app/databuk/inputs/__init__.py
+++ /dev/null
@@ -1,11 +0,0 @@
-import pathlib
-
-# common output driectory
-__script_dir__ = pathlib.Path(__file__).parent
-work_dir = __script_dir__.parent / "workdir"
-work_dir.mkdir(parents=True, exist_ok=True)
-
-# Following is public
-input_dir = __script_dir__.parent / "inputs"
-schema_bukov_yaml = input_dir / "schema_bukov.yaml"
-bukov_locations_csv = input_dir / "bukov_locations.csv"

From b7c8f0ce9b6ad41b85eab0587ba5a5ebedfa7d5c Mon Sep 17 00:00:00 2001
From: Jan Brezina <jan.brezina@tul.cz>
Date: Tue, 17 Feb 2026 01:33:42 +0100
Subject: [PATCH 09/10] Revert databuk/input/__init__.py

---
 app/databuk/inputs/__init__.py | 11 +++++++++++
 1 file changed, 11 insertions(+)
 create mode 100644 app/databuk/inputs/__init__.py

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
new file mode 100644
index 00000000..e122ab74
--- /dev/null
+++ b/app/databuk/inputs/__init__.py
@@ -0,0 +1,11 @@
+import pathlib
+
+# common output driectory
+__script_dir__ = pathlib.Path(__file__).parent
+work_dir = __script_dir__.parent / "workdir"
+work_dir.mkdir(parents=True, exist_ok=True)
+
+# Following is public
+input_dir = __script_dir__.parent / "inputs"
+schema_bukov_yaml = input_dir / "schema_bukov.yaml"
+bukov_locations_csv = input_dir / "bukov_locations.csv"

From e398d0be9e1655a1f71ef2a55553cccd6ddf46bf Mon Sep 17 00:00:00 2001
From: "stepan.moc" <stepan.moc@actis.cz>
Date: Sun, 22 Feb 2026 16:36:27 +0100
Subject: [PATCH 10/10] feat(ingress_server): self review

---
 app/databuk/inputs/__init__.py | 11 -----------
 1 file changed, 11 deletions(-)
 delete mode 100644 app/databuk/inputs/__init__.py

diff --git a/app/databuk/inputs/__init__.py b/app/databuk/inputs/__init__.py
deleted file mode 100644
index e122ab74..00000000
--- a/app/databuk/inputs/__init__.py
+++ /dev/null
@@ -1,11 +0,0 @@
-import pathlib
-
-# common output driectory
-__script_dir__ = pathlib.Path(__file__).parent
-work_dir = __script_dir__.parent / "workdir"
-work_dir.mkdir(parents=True, exist_ok=True)
-
-# Following is public
-input_dir = __script_dir__.parent / "inputs"
-schema_bukov_yaml = input_dir / "schema_bukov.yaml"
-bukov_locations_csv = input_dir / "bukov_locations.csv"