Skip to content

CCDBApi: Fix CCDBDownloader redirect errors #14029

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
davidrohr merged 2 commits into from
Mar 10, 2025
Merged

CCDBApi: Fix CCDBDownloader redirect errors #14029

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
491395d
Fixing CcdbDownloader redirects
TrifleMichael Mar 9, 2025
f40122d
Removing whitespaces
TrifleMichael Mar 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 5 additions & 3 deletions CCDB/include/CCDB/CCDBDownloader.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,7 @@ struct HeaderObjectPair_t {

typedef struct DownloaderRequestData {
std::vector<std::string> hosts;
std::vector<std::string> locations;
std::string path;
long timestamp;
HeaderObjectPair_t hoPair;
Expand Down Expand Up @@ -231,12 +232,13 @@ class CCDBDownloader
std::string prepareRedirectedURL(std::string address, std::string potentialHost) const;

/**
* Returns a vector of possible content locations based on the redirect headers.
* Updates the locations vector with the the locations.
*
* @param baseUrl Content path.
* @param headerMap Map containing response headers.
* @param locations Location list to be updated.
* @param locIndex Index of the next locaiton to be tried.
*/
std::vector<std::string> getLocations(std::multimap<std::string, std::string>* headerMap) const;
void updateLocations(std::multimap<std::string, std::string>* headerMap, std::vector<std::string>* locations, int* locIndex) const;

std::string mUserAgentId = "CCDBDownloader";
/**
Expand Down
53 changes: 38 additions & 15 deletions CCDB/src/CCDBDownloader.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -362,7 +362,7 @@ void CCDBDownloader::tryNewHost(PerformData* performData, CURL* easy_handle)
{
auto requestData = performData->requestData;
std::string newUrl = requestData->hosts.at(performData->hostInd) + "/" + requestData->path + "/" + std::to_string(requestData->timestamp);
LOG(debug) << "Connecting to another host " << newUrl;
LOG(debug) << "Connecting to another host " << newUrl << "\n";
requestData->hoPair.header.clear();
curl_easy_setopt(easy_handle, CURLOPT_URL, newUrl.c_str());
mHandlesToBeAdded.push_back(easy_handle);
Expand All @@ -374,9 +374,11 @@ void CCDBDownloader::getLocalContent(PerformData* performData, std::string& newL
LOG(debug) << "Redirecting to local content " << newLocation << "\n";
if (requestData->localContentCallback(newLocation)) {
contentRetrieved = true;
LOG(debug) << "Local content retrieved succesfully: " << newLocation << " n";
} else {
// Prepare next redirect url
newLocation = getNewLocation(performData, locations);
LOG(debug) << "Failed to retrieve local content: " << newLocation << "\n";
}
}

Expand All @@ -396,15 +398,15 @@ std::string CCDBDownloader::getNewLocation(PerformData* performData, std::vector
void CCDBDownloader::httpRedirect(PerformData* performData, std::string& newLocation, CURL* easy_handle)
{
auto requestData = performData->requestData;
LOG(debug) << "Trying content location " << newLocation;
LOG(debug) << "Trying content location " << newLocation << "\n";
curl_easy_setopt(easy_handle, CURLOPT_URL, newLocation.c_str());
mHandlesToBeAdded.push_back(easy_handle);
}

void CCDBDownloader::followRedirect(PerformData* performData, CURL* easy_handle, std::vector<std::string>& locations, bool& rescheduled, bool& contentRetrieved)
{
std::string newLocation = getNewLocation(performData, locations);
if (newLocation.find("alien:/", 0) != std::string::npos || newLocation.find("file:/", 0) != std::string::npos) {
while (!contentRetrieved && (newLocation.find("alien:/", 0) != std::string::npos || newLocation.find("file:/", 0) != std::string::npos)) {
getLocalContent(performData, newLocation, contentRetrieved, locations);
}
if (!contentRetrieved && newLocation != "") {
Expand Down Expand Up @@ -508,17 +510,17 @@ void CCDBDownloader::transferFinished(CURL* easy_handle, CURLcode curlCode)
std::string currentHost = requestData->hosts[performData->hostInd];
std::string loggingMessage = prepareLogMessage(currentHost, requestData->userAgent, requestData->path, requestData->timestamp, requestData->headers, httpCode);

// Get alternative locations for the same host
auto locations = getLocations(&(requestData->hoPair.header));
// Get new locations based on received headers
updateLocations(&(requestData->hoPair.header), &requestData->locations, &performData->locInd);

// React to received http code
if (200 <= httpCode && httpCode < 400) {
LOG(debug) << loggingMessage;
if (304 == httpCode) {
LOGP(debug, "Object exists but I am not serving it since it's already in your possession");
contentRetrieved = true;
} else if (300 <= httpCode && httpCode < 400 && performData->locInd < locations.size()) {
followRedirect(performData, easy_handle, locations, rescheduled, contentRetrieved);
} else if (300 <= httpCode && httpCode < 400 && performData->locInd < requestData->locations.size()) {
followRedirect(performData, easy_handle, requestData->locations, rescheduled, contentRetrieved);
} else if (200 <= httpCode && httpCode < 300) {
contentRetrieved = true; // Can be overruled by following error check
}
Expand All @@ -531,8 +533,16 @@ void CCDBDownloader::transferFinished(CURL* easy_handle, CURLcode curlCode)
contentRetrieved = false;
}

// Check if content was retrieved, or scheduled to be retrieved
if (!rescheduled && !contentRetrieved && performData->locInd == locations.size()) {
// Check if content was retrieved or scheduled to be retrieved
if (!rescheduled && !contentRetrieved) {
// Current location failed without providing 3xx http code, try next redirect for the same host
if (performData->locInd < requestData->locations.size()) {
followRedirect(performData, easy_handle, requestData->locations, rescheduled, contentRetrieved);
}
}

// Check again because content might have been retrieved or rescheduled via a redirect
if (!rescheduled && !contentRetrieved) {
// Ran out of locations to redirect, try new host
if (++performData->hostInd < requestData->hosts.size()) {
tryNewHost(performData, easy_handle);
Expand Down Expand Up @@ -650,24 +660,37 @@ CURLcode CCDBDownloader::perform(CURL* handle)
return batchBlockingPerform(handleVector).back();
}

std::vector<std::string> CCDBDownloader::getLocations(std::multimap<std::string, std::string>* headerMap) const
void CCDBDownloader::updateLocations(std::multimap<std::string, std::string>* headerMap, std::vector<std::string>* locations, int* locIndex) const
{
std::vector<std::string> locs;
std::vector<std::string> newLocations;

auto iter = headerMap->find("Location");
if (iter != headerMap->end()) {
locs.push_back(iter->second);
auto range = headerMap->equal_range("Location");
for (auto it = range.first; it != range.second; ++it) {
if (std::find(locations->begin(), locations->end(), it->second) == locations->end()) {
if (std::find(newLocations.begin(), newLocations.end(), it->second) == newLocations.end()) {
newLocations.push_back(it->second);
}
}
}
}

// add alternative locations (not yet included)
auto iter2 = headerMap->find("Content-Location");
if (iter2 != headerMap->end()) {
auto range = headerMap->equal_range("Content-Location");
for (auto it = range.first; it != range.second; ++it) {
if (std::find(locs.begin(), locs.end(), it->second) == locs.end()) {
locs.push_back(it->second);
if (std::find(locations->begin(), locations->end(), it->second) == locations->end()) {
if (std::find(newLocations.begin(), newLocations.end(), it->second) == newLocations.end()) {
newLocations.push_back(it->second);
}
}
}
}
return locs;

// Insert location list at the current location index. This assures that the provided locations will be tried first.
locations->insert(locations->begin() + (*locIndex), newLocations.begin(), newLocations.end());
}

std::vector<CURLcode> CCDBDownloader::batchBlockingPerform(std::vector<CURL*> const& handleVector)
Expand Down
37 changes: 33 additions & 4 deletions CCDB/src/CcdbApi.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -667,6 +667,23 @@ size_t header_map_callback(char* buffer, size_t size, size_t nitems, void* userd
}
}
}

// Keep only the first ETag encountered
if (key == "ETag") {
auto cl = headers->find("ETag");
if (cl != headers->end()) {
insert = false;
}
}

// Keep only the first Content-Type encountered
if (key == "Content-Type") {
auto cl = headers->find("Content-Type");
if (cl != headers->end()) {
insert = false;
}
}

if (insert) {
headers->insert(std::make_pair(key, value));
}
Expand Down Expand Up @@ -1971,14 +1988,26 @@ void CcdbApi::vectoredLoadFileToMemory(std::vector<RequestContext>& requestConte
bool CcdbApi::loadLocalContentToMemory(o2::pmr::vector<char>& dest, std::string& url) const
{
if (url.find("alien:/", 0) != std::string::npos) {
loadFileToMemory(dest, url, nullptr); // headers loaded from the file in case of the snapshot reading only
return true;
std::map<std::string, std::string> localHeaders;
loadFileToMemory(dest, url, &localHeaders);
auto it = localHeaders.find("Error");
if (it != localHeaders.end() && it->second == "An error occurred during retrieval") {
return false;
} else {
return true;
}
}
if ((url.find("file:/", 0) != std::string::npos)) {
std::string path = url.substr(7);
if (std::filesystem::exists(path)) {
loadFileToMemory(dest, path, nullptr);
return true;
std::map<std::string, std::string> localHeaders;
loadFileToMemory(dest, url, &localHeaders);
auto it = localHeaders.find("Error");
if (it != localHeaders.end() && it->second == "An error occurred during retrieval") {
return false;
} else {
return true;
}
}
}
return false;
Expand Down