File: virtnbd-nbdkit-plugin

package info (click to toggle)
virtnbdbackup 2.42-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 672 kB
  • sloc: python: 4,426; makefile: 9
file content (252 lines) | stat: -rw-r--r-- 7,388 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
"""
Copyright (C) 2023  Michael Ablassmeier <abi@grinser.de>

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <https://www.gnu.org/licenses/>.
"""

import os
import builtins
import json
import pprint
import nbdkit

API_VERSION = 2

blockMap = None
blockMapFile = None
image = None
debug = "0"
hexdump = "0"

# pylint: disable=global-statement,global-variable-not-assigned,redefined-builtin,too-many-statements,too-many-branches


def config(key, value):
    """Read parameter values, needs to use open function via
    builtins as the python plugin itself defines it again"""
    global blockMap
    global blockMapFile
    global image
    global debug
    global hexdump
    if key == "blockmap":
        blockMapFile = value
        with builtins.open(blockMapFile, "rb") as fh:
            blockMap = json.loads(fh.read())
        return
    if key == "disk":
        image = value
        return
    if key == "debug":
        debug = value
        return
    if key == "hexdump":
        hexdump = value
        return

    raise RuntimeError(f"Unsupported parameter: {key}")


def log(msg):
    """Debug logging"""
    global debug
    if debug == "1":
        nbdkit.debug(msg)


def config_complete():
    """Check if we have all required parameters"""
    global image
    global blockMap
    global blockMapFile
    if image is None or blockMap is None:
        raise RuntimeError(
            "Missing parameter: path to blockmap and disk files required."
        )

    if not os.path.exists(image):
        raise RuntimeError(f"Specified image file: [{image}] does not exist.")
    if not os.path.exists(blockMapFile):
        raise RuntimeError(f"Specified blockmap file: [{blockMapFile}] does not exit.")

    pprint.pprint(blockMap)


def thread_model():
    """nbdkit threading model"""
    return nbdkit.THREAD_MODEL_PARALLEL


def open(_):
    """Open backup files and return FD for each"""
    fd = os.open(image, os.O_RDONLY)
    log(f"File descriptors: {fd}")
    return fd


def close(_):
    """Close"""
    return 1


def get_size(_):
    """Loop through the metadata and calculate the complete
    virtual disk size"""
    global blockMap
    size = 0
    for m in blockMap:
        # use only size from full backup image
        if m["file"] == image:
            size += m["length"]
    log(f"DISK SIZE: {size}")
    return size


def _hexdump(data: bytearray, width: int = 16):
    """Hexdump for debugging, dumps requested blocks in an
    hexdump -C compatible format"""
    zero_count = 0
    skip_mode = False

    for i in range(0, len(data), width):
        chunk = data[i : i + width]

        if all(byte == 0 for byte in chunk):
            zero_count += len(chunk)
            skip_mode = True
            continue  # Skip this block

        if skip_mode:
            log(f"... ({zero_count} zero bytes skipped)")
            zero_count = 0  # Reset counter
            skip_mode = False

        hex_part = " ".join(f"{byte:02X}" for byte in chunk)
        ascii_part = "".join(chr(byte) if 32 <= byte < 127 else "." for byte in chunk)
        log(f"{i:08X}  {hex_part:<{width * 3}}  |{ascii_part}|")

    if zero_count > 0:
        log(f"... ({zero_count} zero bytes skipped)")


def pread(h, buf, offset, _):
    """Return the right data during read operation.

    Function uses the generated block map to check where
    in the stream format the required block offset is to
    be found and maps it accordingly and returns requested
    data.

    There might be situations where this goes really wrong
    and usually this results in an non-readable disk image.
    By using the blockfilter plugin, it "should" work most
    of the time, because the requested block size does
    not span amongst multiple frames within the sparse
    backup format.. as it should match the physical
    sector size .. hopefully
    """
    global blockMap
    global hexdump

    log(f"Handle: {h}")

    # get block where offset sort of matches
    data = bytearray()

    blockListFull = sorted(
        [
            b
            for b in blockMap
            if b["originalOffset"] <= offset < (b["originalOffset"] + b["length"])
            and not b["inc"]
        ],
        key=lambda x: x["originalOffset"],
    )
    log("-------------------------------------------")
    log(f"matching blocklist from full backup: {blockListFull}")
    log("-------------------------------------------")
    if len(blockListFull) == 1:
        log("using first block from blocklist")
        block = blockListFull[0]
    else:
        log("Using block from full backup")
        block = blockListFull[-1]

    log(f"Processing block: {block}")

    # where to read in the stream file
    fileOffset = block["offset"] - block["originalOffset"] + offset
    dataFile = block["file"]

    log(f"READ FROM: {dataFile}")
    log(f"READ AT: {fileOffset}")
    log(f"READ: {len(buf)}")
    log(f"BLOCK LENGTH: {block['length']}")

    if len(buf) <= block["length"]:
        log("Block found contains all required data")
        if block["data"] is False:
            data += b"\0" * len(buf)
        else:
            data += os.pread(h, len(buf), fileOffset)

        buf[:] = data
        if hexdump == "1":
            _hexdump(data)
        return

    remaining = len(buf) - block["length"]
    included = block["length"]
    log(f"Read spans multiple blocks, need to read: {remaining} from next block.")
    log(f"Read available data size {included} from current block.")
    data += os.pread(h, included, fileOffset)

    count = block["count"] + 1
    while len(data) != len(buf):
        log(f"Locate next available block number {count}")
        next_block = [b for b in blockMap if b["count"] == count]
        if not next_block:
            raise RuntimeError("Unable to locate next block")

        next_block = next_block[0]

        log(f"Found next block: {next_block}")
        if remaining >= next_block["length"]:
            log("Next block does not contain all remaining data")
            to_read = next_block["length"]
            count += 1
        else:
            to_read = remaining

        if next_block["data"]:
            log(f"Read {to_read} data size at {next_block['offset']} from this block.")
            data += os.pread(h, to_read, next_block["offset"])
        else:
            log(f"Next block contains zeroes, return {remaining} zeroes")
            data += b"\0" * to_read

        remaining -= to_read
        if remaining == 0:
            log("All requested data read")
        else:
            log(f"{remaining} data left to read, next block to request: {count}")

    if len(data) != len(buf):
        raise RuntimeError(
            f"Unexpected short read from file. Read: {len(data)} requested: {len(buf)}"
        )

    if hexdump == "1":
        _hexdump(data)
    buf[:] = data