wbtiger
/
foundationdb
mirror of https://github.com/apple/foundationdb.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix duplicate file removal for subset version ranges 

Partitioned logs can have strict subset version ranges, which was not properly
handled -- we used to assume overlapping only happens for the same begin
version.
This commit is contained in:
Jingyu Zhou 2020-03-12 15:30:07 -07:00
parent 1a1f572f29
commit 4c75c61f39
2 changed files with 11 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
fdbclient/BackupContainer.actor.cpp
View File

@ -1156,11 +1156,13 @@ public:
// If a log file's progress is not saved, a new log file will be generated // If a log file's progress is not saved, a new log file will be generated
// with the same begin version. So we can have a file that contains a subset // with the same begin version. So we can have a file that contains a subset
// of contents in another log file. // of contents in another log file.
// PRE-CONDITION: logs are already sorted. // PRE-CONDITION: logs are already sorted by (tagId, beginVersion, endVersion).
static std::vector<LogFile> filterDuplicates(const std::vector<LogFile>& logs) { static std::vector<LogFile> filterDuplicates(const std::vector<LogFile>& logs) {
std::vector<LogFile> filtered; std::vector<LogFile> filtered;
int i = 0; int i = 0;
for (int j = 1; j < logs.size(); j++) { for (int j = 1; j < logs.size(); j++) {
if (logs[j].isSubset(logs[i])) continue;
if (!logs[i].isSubset(logs[j])) { if (!logs[i].isSubset(logs[j])) {
filtered.push_back(logs[i]); filtered.push_back(logs[i]);
} }
@ -1196,9 +1198,13 @@ public:
for (int i = 0; i < logs.size(); i++) { for (int i = 0; i < logs.size(); i++) {
ASSERT(logs[i].tagId >= 0 && logs[i].tagId < logs[i].totalTags); ASSERT(logs[i].tagId >= 0 && logs[i].tagId < logs[i].totalTags);
auto& indices = tagIndices[logs[i].tagId]; auto& indices = tagIndices[logs[i].tagId];
// filter out if indices.back() is subset of files[i] // filter out if indices.back() is subset of files[i] or vice versa
if (!indices.empty() && logs[indices.back()].isSubset(logs[i])) { if (!indices.empty()) {
indices.back() = i; if (logs[indices.back()].isSubset(logs[i])) {
indices.back() = i;
} else if (!logs[i].isSubset(logs[indices.back()])) {
indices.push_back(i);
}
} else { } else {
indices.push_back(i); indices.push_back(i);
} }

2
fdbclient/BackupContainer.h
View File

@ -85,7 +85,7 @@ struct LogFile {
// Returns if this log file contains a subset of content of the given file // Returns if this log file contains a subset of content of the given file
// by comparing version range and tag ID. // by comparing version range and tag ID.
bool isSubset(const LogFile& rhs) const { bool isSubset(const LogFile& rhs) const {
return beginVersion == rhs.beginVersion && endVersion <= rhs.endVersion && tagId == rhs.tagId; return beginVersion >= rhs.beginVersion && endVersion <= rhs.endVersion && tagId == rhs.tagId;
} }
std::string toString() const { std::string toString() const {