New v63003/gtm8889 (tests GTM-8889 in V63003) #215
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
nars1
requested changes
Jun 7, 2018
v63003/u_inref/gtm8889.csh
Outdated
| echo "# Terminal Display:" | ||
| echo "" | ||
| # Setting GTM prompt so we can run on previous versions (which default to "GTM>" | ||
| setenv gtm_prompt "YDB>" |
There was a problem hiding this comment.
The comment says GTM prompt but the prompt is set to YDB>. I think line 22 needs to be fixed.
nars1
approved these changes
Jun 8, 2018
JakeYottaDB
pushed a commit
to JakeYottaDB/YottaDBtest
that referenced
this pull request
Jun 29, 2018
…s in YottaDB/YDB#215) The test was doing a SET ^a=1 followed by a HANG 1 and another SET ^a=2 in order to reach jnl_sub_qio_start() while we were not holding crit lock so we would suspend ourselves (due to the white-box test case) and cause another process to issue a JNLPROCSTUCK message. jnl_sub_qio_start() was invoked a lot of times in the above sequence of commands and the first invocation itself resulted in a call to suspsigs_handler() to suspend but it did not suspend because we were holding crit. The invocation that caused the process to suspend was a little later when jnl_sub_qio_start() was invoked through t_end() -> wcs_timer_start() -> jnl_qio_start() -> jnl_sub_qio_start(). But before t_end() invokes wcs_timer_start(), it does a REVERT which now causes a check of whether suspend() was deferred (due to YottaDB#215 code changes) and so we suspend the process right there (because at this point, we have released crit and are out of the phase2 commit logic). And so the process does not ever reach wcs_timer_start() and never gets suspended while holding the jnl buffer qio lock like the test expects causing the test to eventually fail (because no JNLPROCSTUCK message is later seen). To work around this, the sequence of commands done by the test have now been simplified to a SET ^a=1 followed by a VIEW "JNLWAIT". The latter invokes jnl_wait() which call jnl_sub_qio_start() while we do not hold crit and so the process suspends right there recreating the same situation that the test wants for a later JNLPROCSTUCK message.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Tests the following release note:
The ZHELP command does not report errors after the user presses a
<CTRL-C>. Previously, when exiting after a<CTRLC>, the utility reported an UNDEF error and left a GT.M help dump file for analysis. (GTM-8889)