When `num_skip` is supplied to the `Range` header, journal-gatewayd
always returns the very last record even though it should have been
skipped. This is because the `sd_journal_next_skip` always returns
non-zero value on the first call, leading to one iteration of the
`request_reader_entries` returning the last record.
To avoid this unexpected behavior, check that the number of lines we
have skipped by is not lower than the requested skip value. If it is,
then it means there are lines which should not be returned now -
decrement the n_skip counter then and return from the function, closing
the stream if follow flag is not set.
Fixes #37954
if (m->n_skip < 0)
r = sd_journal_previous_skip(m->journal, (uint64_t) -m->n_skip + 1);
- else if (m->n_skip > 0)
+ else if (m->n_skip > 0) {
r = sd_journal_next_skip(m->journal, (uint64_t) m->n_skip + 1);
- else
+ if (r < 0) {
+ log_error_errno(r, "Failed to skip journal entries: %m");
+ return MHD_CONTENT_READER_END_WITH_ERROR;
+ }
+ /* We skipped beyond the end, make sure entries between the cursor and n_skip offset
+ * from it are not returned. */
+ if (r < m->n_skip + 1) {
+ m->n_skip -= r;
+ if (m->follow)
+ return 0;
+ return MHD_CONTENT_READER_END_OF_STREAM;
+ }
+ } else
r = sd_journal_next(m->journal);
if (r < 0) {
--header "Range: entries=:-20:10" \
http://localhost:19531/entries?follow >"$LOG_FILE"
jq -se "length == 10" "$LOG_FILE"
+# Test positive skip beyond the last entry
+curl -LSfs \
+ --header "Accept: application/json" \
+ --header "Range: entries=$TEST_CURSOR:1:1" \
+ http://localhost:19531/entries?SYSLOG_IDENTIFIER="$TEST_TAG" >"$LOG_FILE"
+jq -se "length == 0" "$LOG_FILE"
# Check if the specified cursor refers to an existing entry and return just that entry
curl -LSfs \
--header "Accept: application/json" \