diff options
author | Nikolaus Rath <Nikolaus@rath.org> | 2016-03-09 10:08:31 -0800 |
---|---|---|
committer | Nikolaus Rath <Nikolaus@rath.org> | 2016-03-09 10:08:31 -0800 |
commit | 3caad2930aec3a2af013eb98d9e231f3f6abb93f (patch) | |
tree | c4206545fef08c62473a7f7cdf122e6fc2b73c87 /rst | |
parent | dcdc33cc51b7e4fd4d370df89d5e42377cb04e51 (diff) |
Import s3ql_1.7.orig.tar.bz2
Diffstat (limited to 'rst')
-rw-r--r-- | rst/backends.rst | 11 | ||||
-rw-r--r-- | rst/installation.rst | 21 | ||||
-rw-r--r-- | rst/issues.rst | 22 |
3 files changed, 20 insertions, 34 deletions
diff --git a/rst/backends.rst b/rst/backends.rst index 5bc9521..002793f 100644 --- a/rst/backends.rst +++ b/rst/backends.rst @@ -154,11 +154,11 @@ problems. S3 compatible ============= -S3QL is also able to access other, S3 compatible storage services for -which no specific backend exists. Note that when accessing such -services, only the lowest common denominator of available features can -be used, so it is generally recommended to use a service specific -backend instead. +S3QL is also able to access other, S3 compatible storage services like +OpenStack_ for which no specific backend exists. Note that when +accessing such services, only the lowest common denominator of +available features can be used, so it is generally recommended to use +a service specific backend instead. The storage URL for accessing an arbitrary S3 compatible storage service is :: @@ -173,6 +173,7 @@ to use HTTPS connections. Note, however, that at this point S3QL does not verify the server certificate (cf. `issue 267 <http://code.google.com/p/s3ql/issues/detail?id=267>`_). +.. _OpenStack: http://www.openstack.org/ Local ===== diff --git a/rst/installation.rst b/rst/installation.rst index 0cfa670..6056906 100644 --- a/rst/installation.rst +++ b/rst/installation.rst @@ -34,7 +34,7 @@ that is not the case. possible*. * `Python <http://www.python.org/>`_, version 2.6.6 or newer, but not - Python 3.x. + Python 3.x. Make sure to also install the development headers. * The `PyCrypto++ Python Module <http://pypi.python.org/pypi/pycryptopp>`_. To check if this module @@ -46,13 +46,19 @@ that is not the case. argparse.__version__'`. If argparse is installed, this will print the version number. You need version 1.1 or later. +* `SQLite <http://www.sqlite.org/>`_ version 3.7.0 or newer. SQLite + has to be installed as a *shared library* with development headers. + * The `APSW Python Module <http://code.google.com/p/apsw/>`_. To check which (if any) version of APWS is installed, run the command :: - python -c 'import apsw; print apsw.apswversion(), apsw.sqlitelibversion()' + python -c 'import apsw; print apsw.apswversion()' - If APSW is installed, this should print two version numbers which - both have to be at least 3.7.0. + The printed version number should be at least 3.7.0. Note that APSW + must be linked *dynamically* against SQLite, so you can *not* use + the Ubuntu PPA at + https://launchpad.net/~ubuntu-rogerbinns/+archive/apsw (these + packages are statically linked). * The `PyLibLZMA Python module <http://pypi.python.org/pypi/pyliblzma>`_. To check if this module @@ -64,9 +70,9 @@ that is not the case. <http://code.google.com/p/python-llfuse/>`_. To check if this module is installed, execute `python -c 'import llfuse; print llfuse.__version__'`. This should print a version number. You need at - least version 0.29. + least version 0.36. - Note that earlier S3QL versions shipped with a builtin version of + Note that early S3QL versions shipped with a built-in version of this module. If you are upgrading from such a version, make sure to completely remove the old S3QL version first. @@ -79,7 +85,8 @@ To install S3QL itself, proceed as follows: 1. Download S3QL from http://code.google.com/p/s3ql/downloads/list 2. Unpack it into a folder of your choice -3. Run `python setup.py test` to run a self-test. If this fails, ask +3. Run `python setup.py build` to build S3QL. +4. Run `python setup.py test` to run a self-test. If this fails, ask for help on the `mailing list <http://groups.google.com/group/s3ql>`_ or report a bug in the `issue tracker <http://code.google.com/p/s3ql/issues/list>`_. diff --git a/rst/issues.rst b/rst/issues.rst index ac2cb8c..e79b2e1 100644 --- a/rst/issues.rst +++ b/rst/issues.rst @@ -45,28 +45,6 @@ Known Issues the `--noleaf` option to work correctly on S3QL file systems. This bug has already been fixed in recent find versions. - -* In theory, S3QL is not fully compatible with NFS. Since S3QL does - not support *inode generation numbers*, NFS clients may (once again, - in theory) accidentally read or write the wrong file in the - following situation: - - #. An S3QL file system is exported over NFS - #. NFS client 1 opens a file A - #. Another NFS client 2 (or the server itself) deletes file A (without - client 1 knowing about this) - #. A new file B is created by either of the clients or the server - #. NFS client 1 tries to read or write file A (which has actually already been deleted). - - In this situation it is possible that NFS client 1 actually writes - or reads the newly created file B instead. The chances of this are 1 - to (2^32 - *n*) where *n* is the total number of directory entries - in the S3QL file system (as displayed by `s3qlstat`). - - Luckily enough, as long as you have less than about 2 thousand - million directory entries (2^31), the chances for this are totally - irrelevant and you don't have to worry about it. - * The `umount` and `fusermount -u` commands will *not* block until all data has been uploaded to the backend. (this is a FUSE limitation that will hopefully be removed in the future, see `issue 159 |