-
Notifications
You must be signed in to change notification settings - Fork 1
/
DigitalPreservation-BitBucket.txt
75 lines (40 loc) · 3.04 KB
/
DigitalPreservation-BitBucket.txt
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
SourceURL: [http://tdbm.dss.ca/][1]
Author: anjackson@gmail.com
[tdbm - a simple, high-performance database with nested atomic transactions][2]
http://tdbm.dss.ca/
SourceURL: [http://lists.freebsd.org/pipermail/freebsd-hardware/2004-October/001991.html][3]
Author: anjackson@gmail.com
[sata raid & write cache state][4]
http://lists.freebsd.org/pipermail/freebsd-hardware/2004-October/001991.html
# In Storing 1’s and 0’s, the Question Is $ - New York Times #
SourceURL: [http://www.nytimes.com/2008/04/09/technology/techspecial/09store.html?em&ex=1208059200&en=b00d6fe51d934cb3&ei=5087%0A][5]
Author: anjackson@gmail.com
LISTEN. Do you hear it? The bits are dying.
SourceURL: [http://en.wikipedia.org/wiki/Write_ahead_logging][6]
Author: anjackson@gmail.com
[Write ahead logging - Wikipedia, the free encyclopedia][7]
http://en.wikipedia.org/wiki/Write_ahead_logging
# [#WWJ-113] WWIO needs to flush data to disk - NASA World Wind Issues #
SourceURL: [http://issues.worldwind.arc.nasa.gov/browse/WWJ-113][8]
Author: anjackson@gmail.com
fileChannel.force(true);
# 14.1.3 File Descriptor Operations #
SourceURL: [http://docs.python.org/lib/os-fd-ops.html][9]
Author: anjackson@gmail.com
Force write of file with filedescriptor fd to disk. On Unix, this calls the native fsync() function; on Windows, the MS _commit() function.
If you're starting with a Python file object f, first do `f.flush()`, and then do `os.fsync(f.fileno())`, to ensure that all internal buffers associated with f are written to disk. Availability: Macintosh, Unix, and Windows starting in [2.2.3.][10]
# Failing the fixity check - Adam Farquhar - BL Wiki #
SourceURL: [http://intranet.bl.uk:8080/confluence/display/~adam+farquhar/2008/04/10/Failing+the+fixity+check][11]
Author: anjackson@gmail.com
We can learn some important lessons from this: (1) best data-centre practice calls for independent backups and uninterruptible power supplies (UPS); (2) modern storage systems are complex and may violate some assumptions; (3) because of the caching that takes place in the storage architecture, computing a digest and doing an integrity check directly after writing may be misleading – it can report what is in the cache, not what is on the disk; (4) routine integrity checking is essential; (5) it is important to perform integrity checks on recent content – but not too promptly.
[1]: http://tdbm.dss.ca/
[2]: http://tdbm.dss.ca/
[3]: http://lists.freebsd.org/pipermail/freebsd-hardware/2004-October/001991.html
[4]: http://lists.freebsd.org/pipermail/freebsd-hardware/2004-October/001991.html
[5]: http://www.nytimes.com/2008/04/09/technology/techspecial/09store.html?em&ex=1208059200&en=b00d6fe51d934cb3&ei=5087%0A
[6]: http://en.wikipedia.org/wiki/Write_ahead_logging
[7]: http://en.wikipedia.org/wiki/Write_ahead_logging
[8]: http://issues.worldwind.arc.nasa.gov/browse/WWJ-113
[9]: http://docs.python.org/lib/os-fd-ops.html
[10]: http://2.2.3./
[11]: http://intranet.bl.uk:8080/confluence/display/~adam+farquhar/2008/04/10/Failing+the+fixity+check