Commit Graph

27 Commits

Author SHA1 Message Date
Harshavardhana
ddf3245677 xl/fs: offset and length cannot be negative. (#2121)
Fixes #2119
2016-07-07 01:30:34 -07:00
Krishna Srinivas
5291db60c6 XL/erasure: refactor erasureReadFile. Move parallelRead into a separate function. (#2008) 2016-06-27 13:24:55 -07:00
karthic rao
ce7d5eddbc Misspell warnings fix (#2001) 2016-06-26 22:05:48 -07:00
Harshavardhana
9add048f3c erasure: Add erasure encode/decode unit tests. (#1911)
Fixes #1819
2016-06-26 03:32:49 -07:00
Krishna Srinivas
cb1200a66d XL/erasure-read: read disks in parallel. (#1975)
On read failure, fallback to reads from other
disks also happen in parallel.
2016-06-24 18:02:10 -07:00
Harshavardhana
e8990e42c2 XL: Make allocations simpler avoid redundant allocs. (#1961)
- Reduce 10MiB buffers for loopy calls to use 128KiB.
- start using 128KiB buffer where needed.
2016-06-24 02:06:23 -07:00
Harshavardhana
9b82e64a11 XL/erasure-read: Avoid memory copy, write to writer directly all the dataBlocks. 2016-06-23 02:06:57 +05:30
Krishna Srinivas
d4bea5fbf8 XL/erasure-read: Add Comments and enable bitrot detection. 2016-06-23 00:34:47 +05:30
Krishna Srinivas
17efaaa902 XL/erasure-read: Support parallel reads from disks. 2016-06-23 00:34:47 +05:30
Bala FA
7d757033f2 erasure-readfile: Use chunk size to read from each disk for a block. (#1949)
A block of data is split into data chunk and each data chunk is
written to each disk.  Previously block size was used to read data
chunk which returns corrupted data.

This patch fixes the issue by reading chunk sized data from each disk
and assembles a block.

Fixes #1939
2016-06-20 21:40:10 -07:00
Bala FA
1ea1dba528 erasure-readfile: write to given Writer than returning buffer. (#1910)
Fixes #1889
2016-06-19 13:35:26 -07:00
Harshavardhana
c6ac3fa6db erasure: Fix block index matching.
This patch fixes an important issue of block reconstruction upon block corruption.
2016-06-05 06:19:58 -07:00
Harshavardhana
fb95c1fad3 XL: Bring in some modularity into format verification and healing. (#1832) 2016-06-02 16:34:15 -07:00
Harshavardhana
de21126f7e XL: Re-align the code again. 2016-06-02 01:54:06 -07:00
Harshavardhana
ae311aa53b XL: Cleanup, comments and all the updated functions. (#1830) 2016-06-01 16:43:31 -07:00
Harshavardhana
c493ab5d0d XL: Bring in sha512 checksum support. (#1797) 2016-05-31 20:23:31 -07:00
Bala FA
db2fdbf38d erasure: allocate buffer only for non-nil disk (#1811) 2016-05-31 11:55:50 -07:00
Krishnan Parthasarathi
967c2b2940 Handled possible short writes to httpResponseWriter (#1804)
* XL: Handled possible short writes to httpResponseWriter

* Added tests for Range Header combinations
2016-05-30 11:27:15 -07:00
Harshavardhana
a4a0ea605b XL: Fix GetObject erasure decode issues. (#1793) 2016-05-29 15:38:14 -07:00
Harshavardhana
feb337098d XL: bring in new storage API. (#1780)
Fixes #1771
2016-05-28 16:12:51 -07:00
Harshavardhana
27cc8a6529 erasure: read only dataBlocks if we have enough. (#1776)
Reconstruct with parity blocks if we don't have enough data blocks.
2016-05-28 15:15:53 -07:00
Harshavardhana
d65101a8c8 XL: Implement strided erasure distribution. (#1772)
Strided erasure distribution uses a new randomized
block distribution for each Put operation. This
information is captured inside `xl.json` for subsequent
Get operations.
2016-05-28 15:15:53 -07:00
Harshavardhana
553fdb9211 XL: Bring in support for object versions written during writeQuorum. (#1762)
Erasure is initialized as needed depending on the quorum and onlineDisks.
This way we can manage the quorum at the object layer.
2016-05-28 15:15:53 -07:00
Harshavardhana
a97230dd56 XL/erasure: Reset dataBlocks to reduce the memory usage. (#1749)
Fixes #1748
2016-05-28 15:15:53 -07:00
Harshavardhana
1e393c6c5b XL: Add new metadata for checksum. (#1743) 2016-05-28 15:15:53 -07:00
Krishna Srinivas
b38b9fea79 XL/erasure: fix for skipping 0 padding. (#1737)
Fixes #1736
2016-05-28 15:15:53 -07:00
Harshavardhana
293d246f95 XL/FS: Rewrite in new format. 2016-05-28 15:15:53 -07:00