Metrics screen improvements #1735

str4d · 2016-10-30T23:45:41Z

str4d · 2016-10-30T23:58:42Z

Assigning to 1.0.1 because several of the issues this closes are resulting in user confusion, which we should resolve sooner than later.

str4d · 2016-10-31T18:20:46Z

src/chainparams.cpp

@@ -287,6 +289,7 @@ class CRegTestParams : public CTestNetParams {
 public:
    CRegTestParams() {
        strNetworkID = "regtest";
+        strCurrencyUnits = "REG";


I just picked this as a placeholder; it will probably never actually be used.

str4d · 2016-10-31T18:23:16Z

src/metrics.cpp

+                        mature += subsidy;
+                    }
+                } else {
+                    it = u->erase(it);


The implicit assumption here is that if a block hash disappears either from mapBlockIndex (should only happen on restart, so never relevant here) or chainActive (happens on reorgs), it will never re-appear. I think this is a fair assumption for reorgs, inasmuch as it is far less likely that a second reorg would occur that re-includes a block after the local node has already observed one that orphans it. But if we want to remove that assumption, I can just build two parallel lists (one for tracked blocks, one for tracked orphans) and shuttle block hashes between them.

Check whether iterator should be incremented before calling erase. http://stackoverflow.com/questions/596162/can-you-remove-elements-from-a-stdlist-while-iterating-through-it

According to the specification for std::list::erase, this is fine.

http://en.cppreference.com/w/cpp/language/for

The iterator is incremented at end of the for loop, will that be an issue?

Ah, yes you're right. Switching to a while loop and manually iterating in the non-erasing branch will fix this.

str4d · 2016-10-31T20:20:39Z

Here's what the metrics screen looks like in this branch:

zookozcash · 2016-10-31T20:46:04Z

Beautiful! ☺❤ⓩ

bitcartel · 2016-11-01T16:47:25Z

src/metrics.cpp

 boost::synchronized_value<std::list<std::string>> messageBox;
 boost::synchronized_value<std::string> initMessage;
 bool loaded = false;

+void TrackMinedBlock(uint256 hash)
+{
+    minedBlocks.increment();


Make method synchronized for both of these operations.

bitcartel · 2016-11-01T16:50:47Z

src/metrics.cpp

@@ -121,8 +131,43 @@ int printMetrics(size_t cols, int64_t nStart, bool mining)

        int mined = minedBlocks.get();
        if (mined > 0) {
+            LOCK(cs_main);


arcalinea · 2016-11-02T19:27:03Z

Looks good, but when I accidentally hit "enter" in the window running zcashd, it duplicated the block height line. That line duplicates as many times as you hit enter

str4d · 2016-11-02T19:37:35Z

Yep, that's an artifact of the way the UI is handled (plain console escape codes). I manually move the cursor up the right number of lines and only reprint the info that is refreshed, which means that spurious newlines cause changes like that. The fix would be to either print out the entire console view each time, or use a proper ncurses library. I'm leaning towards the latter, which should hopefully also improve cross-platform support.

bitcartel · 2016-11-02T19:45:55Z

The counters could end up negative when they overflow.

struct AtomicCounter {
    std::atomic<int> value;

extern AtomicCounter transactionsValidated;
extern AtomicCounter ehSolverRuns;
extern AtomicCounter solutionTargetChecks;

How about using uint64_t ?

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot · 2016-11-02T21:33:15Z

💔 Test failed - zcash

str4d · 2016-11-02T21:41:09Z

Test failures are unrelated to this PR:


[----------] 3 tests from wallet_zkeys_tests
[ RUN      ] wallet_zkeys_tests.store_and_load_zkeys
[       OK ] wallet_zkeys_tests.store_and_load_zkeys (0 ms)
[ RUN      ] wallet_zkeys_tests.write_zkey_direct_to_db
gtest/test_wallet_zkeys.cpp:92: Failure
Value of: addrs.size()
  Actual: 2
Expected: 0
[  FAILED  ] wallet_zkeys_tests.write_zkey_direct_to_db (12 ms)
[ RUN      ] wallet_zkeys_tests.write_cryptedzkey_direct_to_db
gtest/test_wallet_zkeys.cpp:167: Failure
Value of: addrs.size()
  Actual: 2
Expected: 0
[  FAILED  ] wallet_zkeys_tests.write_cryptedzkey_direct_to_db (3 ms)
[----------] 3 tests from wallet_zkeys_tests (15 ms total)

@bitcartel this feels like the weird gtest ordering bug again.

str4d · 2016-11-02T21:46:25Z

Moved test_checkblock below test_checktransaction.

@zkbot r+

zkbot · 2016-11-02T21:46:27Z

📌 Commit e80490f has been approved by str4d

zkbot · 2016-11-02T21:46:32Z

⌛ Testing commit e80490f with merge f6b5176...

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot · 2016-11-02T22:24:34Z

💔 Test failed - zcash

bitcartel · 2016-11-02T22:35:12Z

Fix the whitespace:

        gtest/json_test_vectors.cpp \
-        gtest/json_test_vectors.h \
+    gtest/json_test_vectors.h \
        gtest/test_foundersreward.cpp \

str4d · 2016-11-02T22:47:56Z

I really dislike the fact that whitespace issues in build scripts can break tests. I just built this and ran zcash-gtest, and the failing tests passed for me.

@zkbot r+

zkbot · 2016-11-02T22:47:56Z

📌 Commit 6b82011 has been approved by str4d

zkbot · 2016-11-02T22:47:59Z

⌛ Testing commit 6b82011 with merge ae369c4...

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

bitcartel · 2016-11-02T23:00:52Z

If you delete zcash-gtest and make it again, it will fail again (at least that is what I am seeing). After I moveed test_checkblock to the end of the list of tests did it stop crashing.

zkbot · 2016-11-02T23:26:37Z

💔 Test failed - zcash

bitcartel · 2016-11-03T00:38:32Z

ping @str4d @ebfull

ebfull · 2016-11-03T00:39:26Z

The latent slave is still up, @str4d's recommendation was to wait for it.

str4d · 2016-11-03T05:09:43Z

#1718 passed the problematic gtests after letting the latent server be destroyed, implying there is some weird caching error going on. But for now:

@zkbot retry

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot · 2016-11-03T05:23:32Z

⌛ Testing commit 6b82011 with merge 0a1130d...

zkbot · 2016-11-03T05:55:01Z

💔 Test failed - zcash

ebfull · 2016-11-03T14:53:26Z

@zkbot retry

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot · 2016-11-03T14:53:32Z

⌛ Testing commit 6b82011 with merge 11c9be9...

zkbot · 2016-11-03T15:41:08Z

☀️ Test successful - zcash

str4d added 3 commits October 30, 2016 16:37

Track mined blocks to detect and report orphans and mining revenue

d793f94

Part of zcash#1713 Closes zcash#1716

Refresh mining status to detect setgenerate changes

848c89c

Closes zcash#1656

Add network stats to metrics screen

eb5b582

Closes zcash#1688

str4d added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. A-metrics-screen Area: Metrics screen labels Oct 30, 2016

Show mining info once the node has finished loading

24f1026

Closes zcash#1685

str4d added this to the 1.0.1 milestone Oct 30, 2016

str4d mentioned this pull request Oct 31, 2016

Empty balance, 1 mined block #1720

Closed

str4d commented Oct 31, 2016

View reviewed changes

nathan-at-least modified the milestones: 1.0.1, 1.0.2 Oct 31, 2016

nathan-at-least assigned str4d Oct 31, 2016

bitcartel reviewed Nov 1, 2016

View reviewed changes

str4d added 3 commits November 1, 2016 11:59

Improve locking in metrics

cb70983

Fix previous commit

22ee0ef

Ensure that no tracked blocks are skipped during orphan detection

73a4391

zkbot pushed a commit that referenced this pull request Nov 2, 2016

Auto merge of #1735 - str4d:metrics-screen-improvements, r=ebfull

1fb99a3

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

Fix gtest issue introduced into master

e80490f

zkbot pushed a commit that referenced this pull request Nov 2, 2016

Auto merge of #1735 - str4d:metrics-screen-improvements, r=str4d

f6b5176

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

Fix whitespace in Makefile.gtest.include

6b82011

zkbot pushed a commit that referenced this pull request Nov 2, 2016

Auto merge of #1735 - str4d:metrics-screen-improvements, r=str4d

ae369c4

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

bitcartel mentioned this pull request Nov 3, 2016

Properly account for JoinSplit value when deciding if a transaction should be placed in a mined block. #1718

Merged

zkbot pushed a commit that referenced this pull request Nov 3, 2016

Auto merge of #1735 - str4d:metrics-screen-improvements, r=str4d

0a1130d

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot pushed a commit that referenced this pull request Nov 3, 2016

Auto merge of #1735 - str4d:metrics-screen-improvements, r=str4d

11c9be9

Metrics screen improvements Closes #1656, #1685, #1688 and #1716

zkbot merged commit 6b82011 into zcash:master Nov 3, 2016

str4d deleted the metrics-screen-improvements branch November 3, 2016 15:48

Metrics screen improvements #1735

Metrics screen improvements #1735

str4d commented Oct 30, 2016 •

edited

str4d commented Oct 30, 2016

str4d Oct 31, 2016

str4d Oct 31, 2016 •

edited

bitcartel Nov 1, 2016

str4d Nov 1, 2016

bitcartel Nov 2, 2016

str4d Nov 2, 2016

str4d commented Oct 31, 2016

zookozcash commented Oct 31, 2016

bitcartel Nov 1, 2016 •

edited

bitcartel Nov 1, 2016

arcalinea commented Nov 2, 2016

str4d commented Nov 2, 2016

bitcartel commented Nov 2, 2016

zkbot commented Nov 2, 2016

str4d commented Nov 2, 2016

str4d commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 2, 2016 •

edited

str4d commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 3, 2016

ebfull commented Nov 3, 2016

str4d commented Nov 3, 2016

zkbot commented Nov 3, 2016

zkbot commented Nov 3, 2016

ebfull commented Nov 3, 2016

zkbot commented Nov 3, 2016

zkbot commented Nov 3, 2016

Metrics screen improvements #1735

Metrics screen improvements #1735

Conversation

str4d commented Oct 30, 2016 • edited

str4d commented Oct 30, 2016

str4d Oct 31, 2016

Choose a reason for hiding this comment

str4d Oct 31, 2016 • edited

Choose a reason for hiding this comment

bitcartel Nov 1, 2016

Choose a reason for hiding this comment

str4d Nov 1, 2016

Choose a reason for hiding this comment

bitcartel Nov 2, 2016

Choose a reason for hiding this comment

str4d Nov 2, 2016

Choose a reason for hiding this comment

str4d commented Oct 31, 2016

zookozcash commented Oct 31, 2016

bitcartel Nov 1, 2016 • edited

Choose a reason for hiding this comment

bitcartel Nov 1, 2016

Choose a reason for hiding this comment

arcalinea commented Nov 2, 2016

str4d commented Nov 2, 2016

bitcartel commented Nov 2, 2016

zkbot commented Nov 2, 2016

str4d commented Nov 2, 2016

str4d commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 2, 2016 • edited

str4d commented Nov 2, 2016

zkbot commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 2, 2016

zkbot commented Nov 2, 2016

bitcartel commented Nov 3, 2016

ebfull commented Nov 3, 2016

str4d commented Nov 3, 2016

zkbot commented Nov 3, 2016

zkbot commented Nov 3, 2016

ebfull commented Nov 3, 2016

zkbot commented Nov 3, 2016

zkbot commented Nov 3, 2016

str4d commented Oct 30, 2016 •

edited

str4d Oct 31, 2016 •

edited

bitcartel Nov 1, 2016 •

edited

bitcartel commented Nov 2, 2016 •

edited